{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4accbb8a",
   "metadata": {},
   "source": [
    "# Lecture 5 – Edge Detection, Morphological Operations, Corners & Lines\n",
    "\n",
    "**Digital Image Processing · Spring 2026 · Week 5 (Mar 18)**\n",
    "**Instructor:** Dr. Hugo Guillen Ramirez\n",
    "\n",
    "> Follow along by running each cell in order.\n",
    "> Cells marked **✏️ Exercise** contain skeleton code for you to complete."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e925bc6c",
   "metadata": {},
   "source": [
    "---\n",
    "## Setup & Imports\n",
    "Run the two cells below to load all libraries and the test image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67217355-0539-47c2-8109-989f07500a33",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1. Standard library imports\n",
    "import base64\n",
    "import io\n",
    "import math\n",
    "import os\n",
    "import time\n",
    "import timeit\n",
    "import warnings\n",
    "\n",
    "# Global configurations / directory setup\n",
    "warnings.filterwarnings('ignore')\n",
    "os.makedirs('img', exist_ok=True)\n",
    "\n",
    "# 2. Third-party core imports\n",
    "import matplotlib\n",
    "import matplotlib.gridspec as gridspec\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from numpy.lib.stride_tricks import sliding_window_view\n",
    "from PIL import Image\n",
    "from scipy import ndimage\n",
    "from scipy.ndimage import convolve, gaussian_filter, median_filter\n",
    "\n",
    "# 3. scikit-image imports (with graceful fallback)\n",
    "try:\n",
    "    # Base modules\n",
    "    from skimage import color, data, feature, filters, measure, morphology\n",
    "    \n",
    "    # Specific functions and aliases\n",
    "    from skimage import data as skdata\n",
    "    from skimage.data import shepp_logan_phantom\n",
    "    from skimage.draw import disk as draw_disk, line as draw_line_sk, rectangle\n",
    "    from skimage.feature import canny, corner_harris, corner_peaks\n",
    "    from skimage.filters import gaussian, threshold_otsu\n",
    "    from skimage.measure import LineModelND, ransac\n",
    "    from skimage.morphology import diamond, disk, square\n",
    "    from skimage.transform import hough_line, hough_line_peaks, resize\n",
    "    from skimage.transform import resize as sk_resize\n",
    "    from skimage.util import random_noise\n",
    "\n",
    "    HAS_SKIMAGE = True\n",
    "except ImportError:\n",
    "    HAS_SKIMAGE = False\n",
    "    print(\"scikit-image not found — synthetic images will be used.\")\n",
    "\n",
    "plt.rcParams.update({\n",
    "    \"figure.dpi\": 100,\n",
    "    \"axes.titlesize\": 12,\n",
    "    \"axes.labelsize\": 10,\n",
    "    \"font.family\": \"DejaVu Sans\",\n",
    "})\n",
    "\n",
    "# ── Test image ───────────────────────────────────────────────────────────────\n",
    "_B64_IMG = \"iVBORw0KGgoAAAANSUhEUgAAAeQAAAHkCAIAAABgzMuuAAAACXBIWXMAAC4jAAAuIwF4pT92AAM8kElEQVR4nNz9eYyl53Xfidd2932purX3vrDZJGVSlmjLkqhYimxlInsAZ2wF8UyQAWYyg0GSP/JHgMEgGAyQBDNIkADBGEYGCSbjmUnGtjKObEW2SEkkLUqkSEpks5u9VXXte919rVvLoOvT/Orwed97u7pJJ/n9HgLE7Vvvfd/nfZbznPM933PO4H/33/13R0dHiURiZGQkmUy22+2jo6OhoaGjo6ODg4Nut9tut7vd7sDAwMjIyMBxGxwcPDo64jMf+GZwcJAvh4aG+KA/0fQl3+v/zvX2tvq/viwUCqFQqNFobGxs2Gt0Wa9mO+B0xumY+qObDw0Nee/PKw8PD+/v72cymVQqdXR0VK/XS6VS/56oA3bQ7Jv27+fh4eHR0dHh4SH/jEajR0dH+/v7zt14C8Z/ZGQkkUjEYrHh4eHDw8NarVav19X/QCAwPDwcjUZHjtvw8HAkEhkZGTk8PGy324eHh3wfCAQGBwf39/c7nU73uLWO28DAwMEHjY6lUqlkMjk8PDw0NFQul6vVarfbjUQi3jfV2rBz3X/N6L36zCOD4x1PpnJ4eNgZXq1t+j84ODgyMjI0NNRqtXRNIBCIx+OhUOjw8HB/f//w8PDg4IDPgUAgHA4zPppWnkXb39/nh61Wa39/X31rNBrsMtsfOqDXHxkZ4ZqDg4O9vb1Op5NOp0OhkEabl2XDDgwMdLtdrQ27L+xOYQx5BI/TmGgAWWb6CcsgHo+fPn16cHCwUqmsr693u129sp2vcDisURo+bgcHB+12OxQKdbvdoaGhYDAYCAT2jxv94XO32x0ZGWFJR6PRWCyWTCa529DxxGmm1DFtNz63Wq1ms3l0dBQIBJw1wE36CwqvXPJ+tktFG5n1wBRIgGjNa6gZ54mJicPDw0ajUa1WEbbch59rYzJQw8PDwWDw1q1b9zfk0dFRrVaLRqNDQ0ORSKTdbo+MjDQajUqlomc7EtMKU43CQ4eg/zcnabzD4ODg3t7ewJ9Ns7LDfvC9khensU+08f7sGmtUA6gNw47Sxut2u4EPWigUisViwWBwYGCARZBMJpHLGtJEInFwcLC/v681d3h42O12j46Oms3m3nFDIrCvNCaOOLBtb2+PK53lIWnS/+D0bc5B7vtzR/TY73lfbTO2Cv1hnSNN+Of+/j5bhWFEIh8eHsZiMR2QPMu+kRWdrI1gMNjtdpvNZqPR6HQ6Gi67m7w34R0ZQ0Tw0dFRKBTyShwJDkSGbiihbD87z+2lNwwPDzv34QDjcEJ0arU7ehXbU88dGRkZHBwMBoP2wNjb22NUuTOrdGRkJBaLHR4exuPxw8PDSCTC8HaOB63ZbJbL5YODg1qtxgFm16FzZtA3bRbOjIGP1phT51mMp1aUBmRwcJBu6+jVScMcOYPsTIHm5ejoiDNsRG8bDAalU3A6dTodFjdTJW1Fw0GHnGecsPXaZg/9FROvUXiMO/jqZXZBOwPXp58HBweIPLQqJJrvE/v3x/uUXg/16kqSv6i9Oo2ZUwQNmjI9DIfDLNxgMNjpdKRDlcvlvb09DKm9vT0UOv6pFa8FqsXQ3zLodDqoV1ZY2zs4dtVJ5tS5wFnfVg9CEPv+amhoiNXu7VIwGJSFkUgkAoEA3xwcHASDQemAqNWcZwhTZycPDQ0h3w8PDwOBQLPZrFQqnARWXHLE+r4mf+VBSP/BwcFoNCp7ThfzWVf6risdHt617R1Sa1NakT0yMrK3t8dfA4EAhrgzj9yf15Rhx5o8PDwMBoPRaDSZTHLmjYyMtNvtVCplu9rtdlnACLVOp9NsNpPJZK1W29nZYa1yTySYHTG9Dnfw2u6P12S+eP/fBzPwjirfs0+RZvbAcxALtaGhoVQqdV+zZtw12YODg7VardVqRSIRRDmLkmvsfS0G4vTVQUh8B8sKHd8X9v0JZ5FWzONp6N7nevsga9Srxejz/v4++oLW1mNo1l653Edw20HjM/pIIBAYGRnJZDJoKIFAoFqtauegjiE3K5VKp9M5ODhAgiBuJA5YA9ibmO1IeWcGvXoKd3C6ure3Z0WYmv2SFejcDRHmSDH6cHBwwMW9MCLG5KHDa98oHo/rMxJZ559uW61W2QJAQwwUL+JdJ+jdDPvQ0FC32200GqwNq/31V1nUQ5nYw8PD4XDYuzykU/e6lTW0rYrnqPbO0+0hLfwEi43PnB+8uKwTxnBwcBB4LZFIoCvs7++Hw2GQJXSIUCikVadh2djYCAQCrVYrEAjkcjlMmYODg1arVS6X6/V6KBRKJBLCaZ2p7wOxPl7zCmi7LPXuj6R6MoYsM6veeQ16GnDKTwEgzsnh4eFGo4G1xfiyMRhWr+CXIeBV4O0i7rVhemm4viI4HA7H4/FwOAxQYzV924E+Oqn3/k63vfJamoUvJBKLxVCg0AWGh4djsVi9Xre2iPN2UjrsX+k/2tze3h7gg5QgzE+WeCwWCwQCyWRSEJ40l5GRkWq1enR01Gq1EA37+/uoxkwluwKRrQXnbFehIlwQiUS8S1PCyHd4Q8eNPcbseGUr95f+6zgenHF2UIKjo6N2u51IJBgrBkE2vkUAdTcOM66X1qzfClRlWDAxUeis+OPw8y4bWcFeLF6fUYQZE2uUeI8N2n1dZOD+fwcHB51OBwAqEAgAVVnMnUdbKMxXw7BSgDXDwbO3t8fBJvCa9cwa0LAPDw/XarWhoaFwOMz9A4FAOp0+PDzExg2FQkhn+T80L4FAAMGKcAc8aTab6NSdTqdWq7FQW62WY73t7OycP38+Ho8LiGclc1pw5Nhjzwuv2W+8uLNvs5h+n8scY1FfRiIRuXA0jLpYJxPLLxwO62KGqNPpHB0dYf52u91gMJhIJIDRHghrbsoO7Bw3KdoMkC+aZvc5y9rqR3a9fvTzjcZG8nbmsRFw55+OJeuse6+g5zATRoaw4MCXYmIfYTUgaX+IBl2ZSqVardbh4WH4uAWDwVAohCbCkB4cHGQyGQnfo6OjRqOxs7MDpiGND2HNZaDVPIiOyRTw1TT7jFKvpj2m3S5kwDsIJ1kP9ph0dg4jg5KFbY6JrYWRSqW0RDnV7PtqN+pksp1EgjC2vu/uAEHWW+CLP1o50svz6dscL593c3GBFm1/HF8AurXH9U9h94wVb6T77O/vRyKRRCKRTqcx1IAio9EoiiEToacwHfv7+wBrLFrpwq1WCx81HoJOpyNnoMaHOwwMDLRarXg8jhLJjDwGfHrCJu3Bful9nPcCemtHwPe8xFyTXcI3oVCoXC4fHh7iNeQFcSAz4KlkKhwJ53K5B8Ka30uUaN0LBeuFzVvFGbH1UXCJhzZc6nt7e466d8L587IO7J+cQ8jZG/ps/dGMTDAY5DgNBoNCIZkVe/gznmCgKOO6iXSQwcFBBHQ4fH96WO6dTqder6OdIUc2NzelI9tj0lroOGpYEBzaFi21YtpKk/4i23d+ncGn8+IwPCpUJSFiQQArOFAkg8FgNpsFsuBssByeWCymh2J3cwdgPTltdEwyAoybPWMcyNurODu6vB0QpB4vAhyk70+0YgcHBo7uawDMMgaBAw0LQ/AdQ681qUfruKJ7OF10OlrMl4ZMSSaTWNjI1sPDQ9gdEiAo2sJ82u02q/f+mty//x/318VMh1ws1uaQblutVsfGxlDtgXrBEH46CIfHIznk7m7fievTfLWrkzQNl3R5udac+2Ph6UEsj0gkgkCLRCLABmyiaDSaSCTuQ9XHuvL9ewpTFl2Pv7E+hCU5eJBtmmAkQi+x/thHoj0PsOC05fpQNXw72eevXlzY/t8KCzsCqL2J43ZwcNBsNoeGhuLxeK1WC4fDoAH2Kd1uN5FIZDKZeDyO+WJxJLY3L9hut5vN5urqqrQnNBRdbzVK1HmZnGxv64Nl6XN/PbfX+z50rLxfWt8O3QAid9A3e4devhSvn8P6uDhm4AtOTEywJESz092wcni62BSMITJif39fx4mgD3vySeO2fkh1xi6/PsqyFCjrjNIG6TPUgwMPIHuZ/1J+JRGkVPn6J32dVOqDzkItDw489hfXR6NR/XZkZAQdYmhoqN1ui/vEkqY/gUCg0+k0Go12u10ul1mHqOQPXumDIeKH0mzC4TCvxrq1nT88PISHN/yBQ1UiXpcdHh0rGUeutttH7PruZV8s1wuNujPlWRs6F1mW+hKJyjAiKMSTyefzHLpCmcAS0U014A/EP4Y24yUAxc6xlqxXt5cBxaLneweute/caxCt08M5Hvb29iA58LhYLIa3gSNaXCvfSZKUD4VC7FtpN+qqVcp03jirGWMtEolYn0YqlcJPwiDE43HUumKxKGzUyutarYYzQCNGxzgmUUMCgUC5XN7f369Wq41GQ2uI8XHOJ6YJSa0Rs7LGNhlJ0lUtCKChCwQC2LYndNEAVvITuTrRFGBYs8BQHwKBAK7/VqtVr9f39/cTiYTlpep9u90u8mJ/f397exv6YKVSSSQSh4eH9+7dC4fDMzMz9Xo9GAzCXxZJAyMdfRYxbWcNAcrQMTuSX1KIvD5AK5q9e9tB261rCPhbbgltpV5S4L4gPrxvNiHUWLEgP/ZsQGEHTLBHvqN6czZzGkmDlmYQiURCodDY2FgoFAJHVjdisRhKQ6fTQftDX7bUQ8bWUe0ZTOZa2J0XQQaIhxhqHcICUngdjth0Oj06Ojo/Py8fqb1Y94/FYvpe0SFsXk2l1waihUIhUSTlRxE0xMLQdNjNKNcOf7J+6VQqZacDn/Pg4ODo6Gi1WkVWaILov5WZQvPD4fD9MAgtXBE/LKnTWvHqn7PIrFiXo8MXZGAKrQpm15Yz5dbkHBkZyWaz09PTw8Mj943D47cSzZNYHsELtVpNN5HxiAEVjUaZe5HVLYQ3ODgIu5MZtZoFJyFMrGQyiWDFqcjFKAV6NRYf3iHLnGfzIIKRL0i6UCjEMYCPhT+xqy3A2stNevLmKNRCABBVCHEMMbvITnJPbsUssIcHBwcRVQCX8XicL7XY0B14Olqb3WwSNNVqdXNzs1gslkol9S0YDBJugwQU7ikLRlJbE/3QEXPe1Ndl7RXQ3ptIMPX6Yf+fozVLsUByBQP3NRXhNpLOsjWdfgrlwMIT4Y//80/6AArBc7FOaOwvurG5uSkx7VU2WTniukj0eJvzytIVLK3QarV2tYeOcZhWq8UisY4W35ninPAKFntmWF240WjYi3VPTjWdbVJe7cED5Ogl5IRCIQn6/f39er0OywXtgR7ag4o/AX4ODw9XKpXd3V0Os2w2e1/1Q6Fg1p1p8MYl2nGxrlsrea1YkQiwd1PUnIP/WieDejI+Pg7/Aacwz0IhZX0Eg0FhwcPDw9ZrH4lEkLlYZ8D2SHO7o/QsnLMAhXYcOPG63W6xWGw2m0hhvIucbRwGXmPCewIpeMFOdjKZzGQywWCQoCbh0R8Lk9+ZUFn6XmSzj9HXXxlE2WEKpKFIWHMSjI+P7+3tVY8bDm5RXKSMWCT96OioVCp1Op2lpaXbt2/X63W4LhAYIJgeHBwUCgW2ENMh2MHSLfqEyXy8TRLEC6xZtkZ/C13Yo251H82MPMAKRB9ut9s2GFKNzcX+QhxAImIjiPVhKbmNRqNWq7XbbYH1g4ODxWLRshcccr2jSVjhSLMuBy7zhVIdqdJnZDrHNIlmsynD0Vp+fNa2taY5+myf+ZJoUuQOMIMADVkkwMo4h6zzXIwX67cDwxE2zdLF2mPk5fyXcQ+tVvwfGqFYnU5nRLCRHLJYajy4D39Qi4+u2H3CBQpYAh/QD0Fy8UFZYW0tC3RPfis+ZrlcZmPz5tFolANcHmdOfoe+Y1EIxPrAwEC9Xvd9qWaziWNEnEUtKcZUHbMnipjCWmr8HEBABrgMJa4UMV4+cS622q7XyekQ/vqsb19qkRgszv5h9Sj6nLXiBZR8F4PwZf1W4BjAUSQSYeSDwWAymRSdQAtdYy6smbnY2dnZ29u7efPmwsICf8XmIxwZ9fOZZ54h1n9vb4/T2gKvjvHrtfb0+g89kE7YvNyVR/qt/OfWIYzwZe50JT5tDD4rjOD+K/xdaADCDgFhI1EREKD5/EQwlI359HUt6GL71g8lHcvN6Hyj2/r+KhKJJJNJKEBS2riPNoU1awhHsk4Cu4C9PZREAsqzx7wAaIkFqady18l/YH0tnHaM+fDwMItcMM7+/n65XHaoRPb/GmF+dR9eUDQttmo0GtXmgdKotSKIanR01L4w7zk6Ogq9hjFqt9uSO/blOZp4E+uDkpoA4ChPtyWNM2q0YrGoPBUCcKyPYmBgIJPJtFotdL2RkZFKpVIqlQjV811Ddr4dtQUD3HdhMX9eK4H+aOYwCOzi0MVsCdn13M2i6t6uDjxKs+8rXq2oaVJDsB40XwqK1W9BonHNYeggjhkB7JhOpwOY02w2UbR1UDEa8Xgc49pGAFrKh6WQvv/++7du3dKkiEiAA6Db7VYqlS984QuxWIxukJ5FEJavk61X6wUueX/VB9Ow7GlS7nQ6HbLZMNFeQ5uG5WcDGhn8ZDJJpN/IyIgGHzNZlHYMZ9SRzc1N6XSpVGpnZ4dTv9VqMaH8UPMOUifWkMN44d1tJI7jH9L3lvljvxePrZcrxRlquYgPDg4ikUir1eLO7WOoEwxEmQ+wMHRmSIeVJi4xKuYyoiCVSqEdgxERptBsNnnZWCwG1UIPYogY/3K5DGYihIAFzysgWHGzOVlu0N7QZqxrUJOubSi8SxSg+7+NRCIAJZoAcBb0nVarBS8NececiWngsJTAfLXzd3d3Hf9vr7XOn6TtWhsNDynfC7JAkLG2RLTSbtf1TPD+/n46nR4eHt7c3Gw2m+jgvoEedsX0+auvimFxN4dy26vZXSEuVK+H9tEEe4kSGzRkNSMbnuvoktKIBaZrY9M6nQ4BaQMDA8REcALpTBXB3MGOxdlX/6XLc41NcwPfo1qt3rx58969e86Y2HGrVqutVuuVV1759Kc/fe7cOQUf9kLzLD3RV5H5KP4Ap2GKyZaSAWETotkpi0ajOJzT6fTR0VE+ny+VSmRxYQoszQYRhqkn9Q2TxVpja2trlUpF+LWYM3a6HdnqIMWywB66KTBoHDfsycdK97HILyFd5XIZJbJYLNZqtZFhlxVnFSxxpaQewZXiLIedwo5D27U0RKaGARwbG8OzWqvVdAIR1E2OF5Y6ewTZJUtRuLbIFxoWbS59398LItDp/tpApCKj5aWxOmar1SoWi1IWhE1rlQ8NDZHyaXt7+9y5c41Gg6jQer0uzdpxB3uDHh2nhOW9ynkijIy9baO2kEdii+s+NnIEwAiGBnqH77r5iF67XiG/Xtltv+EURN5ZDO5RF33/2C2nKc5CA6v9DCajFa+fp1IpTnfmBcHRbrcbjQYIEiJpcHAwnU6PjIyUSiXNoO2eCDBO4iGty0qlcvfu3aWlJRHyfNcxMWDb29tvv/12PB4fGxuje7IkHLP3o4vg/qigXajMKaOB75ToO4dIoIY1wxrGmQZPRtw4BZ0z4EROKRufmGFkf3RoVKLVimqtLazwE+ftpFnbBdBrDC0ij5By9KH+SH2f0UYUdo5Fk9hsHAzaKdJhAZEI9USYCqRVSDavjAKqVWSjQFHhV1dXs9ksz2VY9vb2mscNNJ/7O/qvVZDVhNojkURNsYFOTu4RK9mY9/t7E6cNByNxyXZYLYjei9sk6x4LtNFo8H/LsrAXW6TVBu95iS/O7OpWQlqsE0n0fsfkB8sWPwHh4iw+74O8nx+qKXsF60OdWronC+JRMwx479OrP7Yn0pQdNijMSzAxYRGcdvZBWrKdTicYDPJPhh3PjChxh4eHVuOwSpyWhIIqbZ5P9sOtW7cwRRuNBvPo5QuTiwq6wnvvvffss8+yaE8iF07IctFY9f+J909KasGSjkQi8GG8xwafd3Z24vF4MBhsHzeGTn4jVA3wolarZSlPoiHKVWu5vXIzOmOisHUr7Bxhbf/q9NY2u2GV79R3iHrNi6UDanLB3FiNw8cKVjqd3t/f393dtTyfdDptTXxgz1wuh+XB8SawjsOS7nGSSZUkNgJSKSfiysrK1NQUaAFepUqlQgCEDZ21JHEbhKlV7Uhekf/s9zZBRS/d7oErhvmWzNWPHS1AS8FOrXKWiouGPg4A5z1XnYl32CbWLHXC0qw/2pq6uokF+AVNMCiky7Av1UfV8hrFfXjc+t4bY+r81hH9JDXExaxD3onCUtiuZoSDypqNaAR2rGhKwES+cg1ILBYjfhdaC09htwhWjsVicBNZglC4rLOOhg3OSW8d1AK1HGtJM8VzLbFSTEdUj1u3btVqNUlw67TRgc2iBa45PDx8//33C4UCaDViS6aoE9jiDQJ48OHw6GjggQYgE1UnjcisbGkREHll9i2yQG4euwbko7bKhKQJ0GI4HN7b24NUDtyvnDyS1DzXCRQCpZSJrdAqoboEuTg5+Rzl1B6oEjoOU+2hCrIXgLIT18uXKIvKEt5JRoZFsnecLTIYDBYKBXLgsGLD4XAmk7FgAOxbQuGtOUJwjfV/MqccjewyXSNRu7y8fP78eXoVCoVA5yRnHHDPRofKD2lzqHIZJpfAWztc3pwkkgMPIhgHHrdp64qDvL+/T8Zeh8Zw8qf08td720k8yGxaoG0rBE/YJSeV5UPDonzpGb74tYUgiL7D4LXOA4fBbUEJJw2uggic+ytEhQxTPAgvOR4q5cFgk+sOOIqdZMFWZjlw1gk1WScriPULKRtct9u9fv361taW1VyU119MUCflAOyI119/PZ/Pk6rbHvy+E2e1Pw4VvZedBaQzN8TyVbpU2MpittjQAWd/gr1q6KToWHoGBh+vZu/mjLOzjPusQ9RqCx5aXcFGV3tJtw6c7RAi+2ck12mhCFL7J2c92AnCyNPaPjg4iEaj4XB4enqauCfYPvHjVq/XGTF5/pVsi5+XSiV5Fy2J1q5e9UEds4lVdZrqJBOa1GfN29HT/S0n1R6TdhF659eFKH2f5zz7oWAfK5gwHrBLS3v4D9jkAiZBRK+ToNfbea3CXu0kQty5lcWI0chsXiHNtJAKhtdBk5T5j0Um8Wp3nZRWeNBgFygUhEpaXMIxRe0assDFwKM0q8myalFOvcOC52NhYaFer0P4Q59ycv56S6IAuYyMjPz4xz/+hV/4BbszvSPP69iIWYub28AcchxCFRBJSSH+ShHOyAg79kb0IYI1Atba1WEMG+HwWLzoZX2Xk2Nf2+bMnYNqesWxUtk41ziQiGPy9m/ifjBBXn+vzl3rzBTdkOgwmJqjo6PdbpekXcPHN+T1UTJQOLa3t3VnO7A2It9r3co0GRkeGRp2/2oVLCWRJgjOiXDu82p2QMBn+mAJjlPXawh+KEXqyZuvq13x3FiyH6OwfiTR4JyTQEsSVfZkOzlb4ySOqZMbAY6wZmKQyBizltp1HLT5II0n6UQAixOJhO6msKahoSFcH1abI5kO6ShDoRBZCFCxt7a25NFWSK62kPg21p39eGLaG/HlKxeOjo52d3evX78uhm8gEIhGo6FQiAh12urqqn6lqDmM/Wg0unLcLl686FsyTfCaZQ3J4ck/7djaz6K7SsHHasbRpz+JA+PVW731FqwcQRwcHt2Xbk6+b6sDeve21RZlhElddXY+PXe8KY4ss5w8h//rrW7TS2BJ6GhsHXamk3tEVdPgMnMu3ud+jIwsLy+riI+sEMEgD3xdg+YVhn4qMcWL1XtZtthJ1rNCDRk3nAr93VGaVqvxWJKYTXbYn05qjYCRPqPsoMz2he0JQFhRoVCwZGTZDicfFN/mcDlPIhO1IZX0ttVq4Q7WcnESnzo7wWLNjqrikPnt+Fh5ZM0Rq0EAdygfG6SLTCaDYpU5bo1GQxFGqVRKXE5ZlIgwlceE6KaaHTK37XDprywO8W31CvJYWKDDq+JZSQH0hmKo/jjruJf2YZNMRSKRXC7XbreJrF1bW6vX61C1iM9Cz4JBFY/HR0dHm81mvV4Hu1cgHxbDwMDAO++8c+bMmdHRUapNkhyHnuBl4vCTGGUHNhoNxUBapzdgMaRv1D2HU4+65whEhw7Ri5ljZV+9XpfnnLzt1v2FngiWLZjehiZYfQ1I7UEkxQcJux0vjraJd4rlsdRCsndwTG3L59GfOA7F0db72ix9+PeUbw/RXKvV8FHjPgHKUJ8HDDkPMS1y98GhcYMf3O+MfIDW3XrCOidWQ9/Z2ZmYmEBpUIlauO19fo4ktPE40P5oStgkv4tq7MnwsnP6QCEeePTmtesxrp1rHuPOvR73SHfjABRfR4NiueGOvO5lKup6mwfZS85jL8mp5Uh/G6BB6k6VpMMhhqQTqQhaIR4qshfBZgEqZWMUi0UsawcLs/23X0pHRrYq/UsvjKsXT9xeiVWo1NXeir22D75liHlKu92en59PpVLZbBbvkOraCLeVraY8efjruIbbcuXe3t7Ozs7a2tr09HQ8Hs9kMhZTlihhPQgWZ3wID5G5Cu6JoxW6mH13u96s53PgUZpwZHs0wkeEnabEA6otK5CBjW1XsuZd0Ua+zhI7/l7V2ztZEr7OD4UwcK44XH6bc1UtGo2ypHkXnYW+wJFSitu0ooPHz+3ude+nXD2Wa3LO6+dyxTvcMy8ifMKm61ntXl+O92JfVVVnmMLZpfirVpfIdVbQMwKPgFn3ecNQKGTPz/8YGvFjZA5TlQoNhN0bvTikkiYsPjnr7Kbltiw4jiuv9mGFtSAjpod0VCqTwdIHjCO02ro77KHlwMeO19RrWLFK2Pko2rKBvBvVETq9LCTeAmFHeN7JRZVjtBGh8Pbbb7/66qvPPPPM0tIScRDovCrSCt0KXFgh+3bzIFb29vZWV1dHR0dJcsDhRxNRFUEATiqBi3vcRuI5gJUTVuO8zkcxH3VwikxCP1Wb2BaPf8AN+CCDsY2t4LhSYtheI28xa/snqcNOmQInA6WlaSs00Y4J4SQomFYDbR03jknRV2zgm5O7RgkCj475MIlEQv5YsI4PlRwaHDoa+Cm/RTd0MOKTS2oH6lGaDpuMwbtTLI3dIaoqxZs9/HCTwqQkQpLIJguAPIA6T9jjXuYDa4Xl5cRw9/LmOa3PZQ46cUJPIA2lD5OK11biCALr7Q+9WoPzrGg0qnwmNs8JSiXi1XqlrQy1i5XsfQi4drtdLBahDfFPiXI5qXplNvBWhLCOdS+7U9xSVSC1SVDtAPap4+c0qFFSjqxd7Os86TVrJDxrNptLS0s7Ozv1ej2bzX7qU5/a2NjQAbm9va0EOuLj94r0GRoaunnz5tLSEqwvh4SnXA3dbpedI3yMvLve2jH2sLTHp6KxThK75B1huqSYY1VQFfqhiGRLa9O7sIGtRJZ85xgTuciOlTyK6rbgVGmpNNF2fRO0WbzVasH8lu+BtuwYEniJpgIBEe3HSSmhFG/W+z38QVIHGbhgQQ98LYNDx3XQPiSjfUlc/c9Ux+nqXCwAzXfG+RKM3mHLWK6qzfN1dHSUy+UEjeJOJzEZse80BPr9X+JUVVIkpd8jRtPBiy2HjAc0m02GXhkyfY9rX1zYSmG7o4ScWkhBAkLB/k6UjbDUSCQSi8UAhoLB4OnTpy1zgJwhXrjDmU67V6PRaK1WQ1lgt+MPoQw8mQ99ZYcDltFs7RJrslkBZ2fBW0yWjaGkg/ac9wUi5DxEeXQgC8fx7YVBNRo4JKlhAceGxeNo3741CW0Arl1UqLeNRmNtbQ0FamhoaGdn58KFC4FAgIKWFL6hwtPOzg7+JfHk4G9A8+ehrVbre9/73s///M+zPXwBVkyivb29ZDJJhQ6yb1uBZXksloHj7GGvve94EXt5BTXIOuHEUxR+rcGUOsbp6DCXYT3WarVoNIpXmdB2sDIcDDbbp2+QlAp72sxzTpYrJ9+9UBdY4Rpt6sVgklo6uUxbST2vp0cKlnhy4Q+4ych38H2WMdCEdWDaXLJOxRbtxF5atrUtuIDUpmDWcOGbzaZMPZ5lrQGb1g3/E8oy2h4cJ/wfSEuK/9ocG6huONtEk78/dApmtVHdrANVpXJ2viQ1p8T4+DgMJ3nAreCwiqpjTDn/t6JTTiHBcxp0lUryspf4TDUs9jNYpC1/5WgHTnxEr6KoJMchz6RWKsQMEAwnS19/c4EPThlZKxw1bpZ7JEsKU8Z+6Sujvd9I31G5BkdMP9SCicfj7BBWmwSKAw7aveHdDEJaeBzpxdfW1orFImHWBPKxAmdmZu7evRsMBklOq5elALZ9onKIw8dYWlr62Z/9Wcf35ZhQ9ntFbPvO2sfSfM1Ep2M6SGyTy1dSzEaZ2uIpgUCABGeyd1W1SxnQHEaE0z1hGiqv7AygJdrbiGJVL+IyHC26ea/jnyalUHlybKZ/CyMMHDcHlbIqnSS1FwmwMtraRl7kxB6ThDJpgYVCIRXFtiYORauVE5TfWqgwmUziRUPdIWyQ3Ms4GGq1Grry1tYWRuHGxgb13Uul0rlz5x5EXknhBysQlSKZTOqdVcVABGGuJDslGlYul1NwkePBcybJK6BtU44VgDkdTSq+Kf3au+aWl5etfqfz1nJ1wfWsxufV951ecR6Ch6Cw1Ot1ouwkQ3VxfyaMfWuvUmadD46G5auUOVLeIep4n2uPsV60Ice5r/9nMhnMCOSmOAOOe8fLjnBmHCoeb0RoyU9+8hMqhBYKBVIXEW18cHDw/PPP7+7u3rlz59KlS81mc319vXHcnEAPYsx0imwctzNnzvjmn7RYjZULvgNyklx9fVovRNtLb9csg2CIPKDvlaoMKWyTF8pogCdD0SJlDI9Go+l02jrAFf7jmLxKsOUtoOUQEG3gD+JGMkvCutfgeLe8r37gjemlKYDeutlpXuKAL2ZtVSXfzuh4yGQyXIxEKhQKqVQKz7OV7yjONp87i1OMFAFK8lcdHByQVQlJXS6XQVlBBWdmZsbGxtbW1igrc/fu3fspUoFiUbbD4TC8Qr60+XMVM2YLAPJ/Zd+36FUvJxWnurWwLKOOa+QhpfKLpbkQ8QmSwMX2twITpJJYx4sznfZ09e5n54A5ODiggpRyBRBDZalg3uaV4w7QwTc2W6mztnSB7aRNZOPc3I68U5JKwyV7zYZx90/Eyh2oQkTcEw46HHq2P/b8IHs191GaTbvEiVk/PDwkM0Y2m8WUTqfTyWQyfdwajUYmk3n++efff//9Vqs1MzNTLpeLxWKlUrG6m/LSib94/fr1fD6vanBOIRLbCONEMxh4xHYSh1Uvao09POx9bHU9Oy8MOMVf+DkycWdnRwnUcrkcNrs4hZjbGCI2itJ7jEmzVq+sMmHp9o7rTJXLQbHwkPcZLkfDgFXtWIrSZrzCevgYcJeT+af3HBgkYcBDgWmvyugFA+hPPp+Hu8n2x189NTVVLBZ1JZE7hOBpISlRquqE6e0UraoQ4lAoND4+XiqVwuFwNptdXFy8d+8e99ne3j579uzq6uoDFJ/coQxZOp1ut9s2MaBXAUSb5jMJIiTErfmmnzjuJk05o8DSVNCKtK1kMkmRQ14PAbG1tUU5TrAwB0+0FFH76F7+Ii8X2FfZZ+FygEHYwHqw1RT7/Ny5lexEmZyO1cbPEYKOy857fzuezrvYyGNsLuWblo2mK62uIUeorFqkxs2bNxXCI8QAo0fnN3LEy7FV8nFWsKZgaGhoeXl5YWFhdHSUEc7n8xcuXAgGg9vb22tra/gwQ6HQqVOn5ufnV1ZWzp07l8vlFhYW4vE45S6LxSJOF6V/2d/fn5ubO3/+/NmzZ+m/uLF2cBRorhQfVl57LcJeCmCv6da8WNvIrlsxX1WnUVCkTlMsS6iKkpitVqtUKq2trd28ebNWq2lNEuw3Ojo6MTHRbDYzmQy1/tiwXtux1za3GRFw1GtHW+ayHA/IaMVMquqjlihsB2bTlnxyQtttQjfLwzkySIhcUGj0wtnxn3E42ah9pcJ3lCpBrE4yeoA4OQ+U3h0HCTZHKpViK4EsUwWJvaNXEBNJKZzYaKQX1pdoGPl8vtls1mq10dHRmZmZjY0NNinL4D4MApnJQnsUL3Cy3znOd1tpRYnM+aeQNfFbtT4cVNfCcM4qEaWcIVPq6nK5rPgOq1Bbc9jRGb3bzInr6bXHnD9pPcl6EAhu5Wyv2/rSpKyn2GkybO1G8i0n4b2n94NtKFmSCLYb1g1oZ1yWI6Htdg/o7Wz6TXEELdICYVy1K/WaAwMDW1tbLINKpZLL5cign0gkxsbGzpw5w8GGU+Ts2bPLy8uLi4tYV2yPoaGhVColzVETsbe3t7i4eObMGZuu6PGyzvZp3rl2fOPOZFn1wjr0gBn5EvxHxTSU6hPyQ7PZrFarc3Nzd+7c2draQkhpIiqVCtFArVbrwoULnI7VarVXXURn9XotvF7NweXsUMix5AyR5REoFZeTfVNHOEOksIlBc4wREQMAAmLG6iVhmcr1yq8mGMCm84aw4NRjopFbDR8+tQBtdhfWNiOM3uOEuWsQVJTAKQ/GP5HdkHRxNqjGSyqVIm6TtbG1tTU1NfWgFILjGJT4s9Rd8SudZWqLbMIp5vBXWULLY/X6VWxmLycwV4sVRalWq+3s7HASWhXjJL417zo7yVq0znpr7tmSNCexgh29QE0qszX9uEZYpBaTgg68QZje1scShHrxU3bqB/93CpTIYeL81sogZxIVlslGshMqM9kud94F0p784MQcE71JAoNWq5XJZDY3N0m0nUwmL1++fO3atUgkQqU3/CXeETg8PFxYWGg2m9lsFlUO89mRR76BTs6t+s+s93pLQxYE7NhGR0dHqVQKlBOuuuUVqHG4UkQ0Fottb2/v7Oysr6+//fbbAhwk9FHDUbRHRkampqbGx8cxF0RhtCi5dzM6CZs0TY66Y9eM1clEKrVmioUW9UT91rLFdaV131nRP3y8CxSEJQgYRSGfz0v/g6cvVpVCS+zW0Hp2jnBbxohBVhkEaVdK6mJtaysxhPiLMSxyi+WYo5UPDw+vr6+T3uDu3buZTMYWRWq1WgsLCx/KUWmPRy+3wTF87AbWLKrulwrleUExr+7p9UMKs2a+lYqh0WjgObEWlvc+vo61Xn89yT7UBNi1pRDtEzZr5Frc3NsfOXmkmdr9z+p5pJ7bz72CZS1/HKanYzaCLSrhvT2k0T6YFA4z9IVe42DLFzUaDcg2R0dHk5OTeM+SyWS3293c3MSGm5ub29/fv3bt2vDwcCaTeeKJJ65cuVKv1yuVCiqJMBDLgCSX5ubmpqBziS0n/4aDoj7eanHGPxKJOAivvYBOqjiAqpDgM8TvRKtUKhJqu7u7pVLp3r17d+7cYRkMDQ0R9mnFnKJIyFRnU636Hjxe09D7V9895TTHqywCrt1BFqa3oIr132jB22wnQx9QTdAkCP0NhULFYhEuKTC9onPFgHDqillBZGsB2s4TZQ6t0+batnC/fS+HQCzICAeMyIVeu3l/fx86MpSPfD4/Pj5eLpcxYcmd3Wg0ksnk/de3Obe8QtlBtZyoJ5uz1U4S13s3g/efVjrblRGJRPBioQrV63WQINSo/hyGPnh0n3Xm7ZWzgnE12NSgvoyX/nvbCRixJPSRkZGZmZmFhQXLSFMkmx1kBzWSsg+dVuVXZAbiA8Sy4yRPpVLscBK3O6gl/weyUIZfGj4TtFTLtG00GtFotFAogFfgerXJQp1xsPYTabUxZpUqGsKJaIK7u7uNRmNqaurUqVMUSVlbW/vKV77y6quvLiws8F7gJFZesyXm5+fx4KPAylbDNZJIJMj0tre3V6lUHPjS4vVODn6pFFb7Y7SJr0FLkvWZSCQODw+LxWImk2FUAYL39vZUuVTRsNTuEKzUarXS6XSz2dzY2Lh58+b6+jrvKJcAMki1T9lNZFXELZbP51XeU+8lKeZ4hp296Zxhjg/cUcKsv9RG4jhQuAjjoMA43LyCj/Wpwy/wQdV2xpb7z8zMSGOVCoXv5P5TjsNkhkcekBGJDNBL4atTYJpzLA0ODlLuyisJ+8R8cX9l5lOVL3sZT+QzjwBtX1paIt9ktVoVtMXLPig+8Ki6wwkN//4/6e+IwySUX0iRcvz1hEUOH6PbvlGkdkE7ZKBe79VHu7cWqN0YGCi+VU6YV1sl3et1tNWXhdBJlESjUXwm4oFJDlr9xe436JhICvXEWve+Tk78DVJt+uhfetbCwgIdGx8fR3NH8FHXmFQS+Xx+dnY2Go2yBzqdTqPRuHPnzjPPPBOLxbrd7sbGhjLwyUJnMFdWVqhdZyPLGSuQGcWs+y4JNF8rGcUL5lZIHF4WTYqzB+VdvAgMaooFa4TRmmWoqTSPtz7LznG7cePG5uYmRF2rDFrMV8gDOVe73S6MGsdlYjUw72Lur2s7Gp6XfSRd2HGrkngLX6giMKvVKo5lZgEEAGtAWu3A8c1xkkMNVKMKF/xlsolhlKsyBlsAJ4dKCTtnia8x6iumfcfHGQR98OWJiYzrABV0mzVjb6K/+gvrPsLo42qOxeT8Fe8/eddYGfhV+jNebfNe6YsMeJszc76y7PHGynss25N5d3eXYFw5bH05hdbE4Z9Ko2O1PCll1gfLYnUq2Fqvjk57QT1O0iJbNco6Oey724K8vkvLSaOIToGmiUQmgpGdjH9sZ2fnypUrly5dSiaTgUBge3u7Wq1WKpXbt29fvHiR1HqLi4soIwKRuH+xWFxeXi4UCvV6PZPJCEGmFjuIMAo1681WR7LBX9ZjRmwUgh4xoSBpViyjTeeRnmRwjcViqmALl0C/RdtytPijo6NKpRIKhTY3N3/yk5/wW2bcWgAOJ1pGTzAYzGQyxWJRuV7tSemLitjcIA6KbSEsL7/L6hDKM6XURVJdYT7AFYbrhYpN1HHiuAkzsUSGeDyOlgr+Y2N2Go1GsVhktEulUjAYHBsbU06ewcFBKu0qaaUWfC8B/dB2EjOa89L50gk1cs5LnDRKNGS1sQfCeuA/voZxhLWOmgYGakOBH7VZAeF4Ux/6K68r5jEskl7aCrdCWbPz5Mhom5pVlVOUIF/YnOIpHuQn+yBHnQIXEabyzzipMLzcbSfs2Kba8RYcceAvBx9zBkTOxr29vUKhkMvlIpEIuAQbD42J8dna2qK20+LiYqvVSqVS1Wo1FArdvXv3k5/85PLyMjC3tQPU7t279+STT9ZqtWQyqQWQTqc5GHgXon5tdXC5dhUEq9dhNTKkePxl2aBZk8NEI8P1BwcHlGFE4nABjlxRGtDHZVZyzcLCwptvvokZBKuVEdAaUKoyGxgxMDBASSpcFDZtqV14VsX2ZpuxrY+XxV6gTHKS+KxAGAdEKjBcFC0CQwNbOzw83NzchHsnl6lu3jmG5lDdCB4mLFtGCc5zVaTU5q3VaswsppsYIxYsPUnr9eInpBhZIpluZavCc4bJGnB+7uMf62W6ei0mX8imz+nU63vnPqQSJeU8FgEXqKhjr+YddMds0QCxN2wueV9kRuoqpG9tXbSwB9ni/WIpHSaivue57FsJDmmyQ0NDhEvJnSJhKqKFGN82S4O8Sc5issknrQccKUM0kwNZ+obGeDUsiTwbdoxbjCkDPbfPtTzuoaEh6MAbGxvkT8CFwvtGo9GxsTEoEAqlI3fH0NBQoVAgWPHcuXM7OzvDw8PFYvGrX/3qN77xDcG+tv+dTqdcLtdqtcuXL4+OjpKN4ODgQLX7iHLSqWB/K0zTYVjacBU7IE6+JI0bUW21Wi2Xy8kxXqvVdOja6q7AILVajXQo7733Hp4MxDQPJR2Ks6rR0AWzTE9Pz8/Pj4+P8xTHf2BNtz6Ki2MB22oSzo5DOObzeQBohpSqklYT5wC2lXpisVij0eBEQRYrRFYzcvRBpAUr1soijnB+Xq1Wk8kkiVhhIrRarUqlAuInWr3u4CTV8pVsvuiQgynZfzLaNixcI+aNcrL2DfvR+iHsI/7j1aw7nQ5uB0yJj95JR9CAxtpMxLqMDzxR6XJSqRTSWVwCTn4pLKqIYbcrn61GjCqnfD1q5KahjDcKBfeXsdxut5WjQKk+mXuEncoq6lnWUWMHUB42AERiZG2KLzsU/Tew1ctsmJw6YNPFONxEjh+MPjnxSDkG8IIjkZkSFtFsNnd2dgYGBshYnc1mI5HId77znV/6pV/6zGc+8+KLL5IU10KBnKlvvvnm9PQ0hyWjpOLW5IRRmm9vtmKLuvbP1qTzyUpqp4YAApc4l14aGTAguPy9e/dqtRqHt95IdXudOGYtGJJLqPSoFUCWj+E8V1qwtb3svEsZ9xVk8lIoWhVZbFOoY3ZYIsPBwUGtVhOVTY5Qx+N9dHy9U2vUdgxZLHcFg+ZlJdlUNr0Wdq/Wi6pg953toTaRxR5tUIt6YtFLxxuk//szz5yQhz+75mVtazRJcddnQfdvvcA4B1jQLvK6yyBXQPsNBoPELDSbTam6kkFIB2eBejVrKZWsaVvmg8RvClja29ur1WoHBwelUknhKlNTU/F4vN1uQ7kVTClnju+J7X07nouvhhwdbGlfBNPX9+KbdSRy3BDWslTsQeWsKJTNVqulwK2Dg4PR0VGKew0NDVWrVVVYJwEkxXCvXr06ODh469atSCQCs+LixYs3b94cGRn5/Oc///rrr5PvzW6JTqezsLCwuLg4PT2tnYMDymZoE0ZkxWuvFdLHjHPktaOOyaCJx+O1Ws2zcn9KKbt79+6tW7fIgqL4cgpCoqLqiZZ2OTMzk0wmFxcXb968OTk56dUQ+1SWAMZxjlXnGoWwOgAauZhFKNYJx3kMygxPmfuj6lp9U3i9k8fRSm01XwqA9FnCZdngTtI36wqy0JC9fy+Z40yxhTL0vYoF6xvBaNaatwetHB5OCdZ+wroXrPF47STqsB0U69qORCLUqI7FYsoN4sQl9r+b7zFgLSANkL3SLnp44ux5FiLFRKQdKwxXsQy6lTWZ5dGWHcrqtKhFPB4naQvasY5fy0cuFouTk5P4ta15aD87ya+d2ol8VjLbcDg8NTW1uLiIDuIdK98v9Y1zc9Q967nyNS11ZyKbFNPIUoajub29jalhT6NGo0FMVyKRmJ2dvXjxYqPRWF1djcViV69evXvcut3uM888k8vlXnvtNdHz8QSm0+l33313ZmZGHkjMZNthYYW9rOCHLjzvD70xJuJrUgdWd9DCAK9fWlp699139/b2MOlkpVn9WkeIhRrICvTjH/+YskR22dtOehtDrQIrkUjEu3cU4kGOAfs9VhS1BbREFV0pYrtWpnXkKmBHHmzHtzbowSIcSMeptaj80VzAoeirjnws4q6Xsi/Hg3Pw2/fyjRZ2WCIfou75imkbDmuZ1wrWoFmw3LvVvcLCqhsMMYQnOZRGRkbGx8fpFbheJBIZGxtLp9PKZa5J4ulevrB6otQiUOhZ9zZeHAHttbzQ6XQYIlwgmSnPjkhXzoZUjQ/rkLQmknKecT15VqHcomLzPTF7qFHJZDISiZRKJTpmTTCbTlcT4c3AJxcWLIharba4uIjy7mBw6rb1PnlTG1PSEDVfzF9bgwbB56TZtPYE31ALOJvNwjlNpVLvvvtuKpUaGxvjGhBM1iGevcPDw0KhcOHCBWDl5557bnp6+s0331xYWDh16tTZs2dXVlZEEQsGg+DdGxsbk5OTOqodwkwf5KdXcwSHhsv6aaE/AstSqZ1YFUhmukkoFAKNCQaDd+7cef/991HAVdoK2A0VAZwdjiDp9AYHB0ulUqFQCAaD165dA16TouqlBgMoK+6Zpc4+0lpVeLR9ZR1mqMaaULFZLL6vcpFWa7b5pJgdjZstquu1Sp3PToC05d0i/eEjKEkIF+tzf3a5Yw95T277WXCfTUqhTlqLzR63ysaHpyGRSLBcRSdH+BBg+CAoxtLjve+jJpoRE6BNaBkFVkBb08DmDyFbGwsXuCORSJBgKBwOI0eQIPwWeA4zKpfL6Z56MQeBAqpT9kgb409gHnFBGllIuPzWVntpNptOBmFB2I4QlLC2uoBXye0VagTGxwgoma9FFTXCqhM40KN5rXV7VHAWyumMYutNkehdi76NFYZvE3mh4oHqDLvRm4EhGo3C41b5VCAR/hqLxSYnJ6enpzOZDMcV61NWDmSvb37zm1iOY2Njp06dKhQKn/vc5/7tv/23RHyRbozAa1ZdpVKZn58/f/58pVJBbXSOqEeV1L1gIv1JUy82HonWODLh/OnYRstuNpvvv//+jRs3cMAqgpxGdmNY841GI51Oy1V+eHiYSqWmp6evX79Obc/d3d18Pk9xu1AohHy3kI7S5OIpsdHY0viYNeuKAAW2Of51XPVJhOALqjgX9BreE87CgCkr3qsQsNfI6AUAPrSrXvXOUUYlIW2yUq5RwK3GGdUESaV8T7ICkWn306cqFT3N0uNw2iLmZ2dnlYlNUcWID13PHrCv5IUj5N2S25RrCPFqNBpk59EIcmJ7xZDKSqmWsC4AI8OEFHKKs5h0Qq1WS1JYQUreKfFCk04aI5sf3bsCvG4cu7dVZBrFf3V1lSFF01dKWLnXkGiI1z5wUK/1bUuEKDG840HyXdy9HN8K+o/H46IP4t/TBegRosfaHypQk9OaQ6hSqVCRZ2Jigkox3W63Uqns7u4qP8PY2Nji4mIoFMpkMpFIBApto9H4xje+EQ6HL1269Morr3ztN75248YNTHJ7ar7//vtf/vKXYUk6sQwfi6fdpjm0Jz1TpswktjqX3di1Wu327dtk0ZOVKXof4ANMJFjJ4+PjY2Nj5L3iCLxx40YsFltbWyMxw9TUVDabDQQCRGM7HcNYkVC2whf6hO8CluhxSi9ycnvHxObPsgrsYwz44IfVXjt3Nq1rL6zDV8j2Qef7zDJcF33jlFgSAGJ3uoTn0NBQNpu1qfaVW1Epq9jm8odhI47Mzs42Gg2rBYtMboMC1A8VmQadcAAXmy6VyeadJRoEb2HdY/WrzLtGzZLh7X28yepYZDr5rVUra+tB4Ong4ObmJowLG0RkrQq7GvRXZ2E5Gr1WiSOvfWMRe7V2u51OpynIxqJnfWuJKy2Ow5k/YQN/xyPv9K2PX9H7mWaFeDwez+fzNpeQo9QIEXJObii3LLZ0Ok2hpt3d3WKxODY2hiwGAGm324lEQhSCTqcDU2JycrLZbJbLZY7/QqGwsbFx/fr1TCbz6p+++iu/8iu/93u/xw+l4dbr9e3tbRQUyzKSfWnt95MMrPc8k6YprUqaNfdXCgu9ka65efPmO++8g6oh/6EKUQ4PD5PvGOAiFovduXNnbm4uFoudOXPm6tWr8/PzTzzxxLVr1/b398mdaWvuiNLnYKYihnOC2jokiuCw1Vu0IAUkeg2U/sP1MTrG+jSv56DXNY+hwjuvgKOeRkESdq7YtwI3BHGIDENx4UajQTiPvJHWv/VgzOGEJRIJEk4SJWjhba8PkO2HbTUxMQHeGggEyDTdH2aSYqUSFTKdVFLeaq/eyHJv4XoHsaLxW5YUa2trawtkg4Rk3oB97wlpgQv2GylXxLT1GlbeiaT5VicAzEqlUiMjIySQU5IEu6PC4XAulwuFQqR68JZk9H0FPYJjFSeeUm46SLqSS2jo7LFHFIOCyNHyMKSwh9LptOJEOAu5WBaPxRnFhAH2SSaTzz777Hvvvbe7u0ueXyItifuYmJgIBAKZTIaci+VyeWtrq1AonDlzBrgcib+7u7u+vg5hZnNzc319fWFh4cqVKzdv3rQjEwqF1tfXU6kUUcs2IJ6OgdjQ815pHZ3Bd1S5n/oShoYPDn+ah4vTHSJtPB7HbUhJ2ZGRkfX19bfeemt+fh7TljwYtniVdorImrjfA4EA2sN77713cHBAlsFCoaDStLu7u3jpVSGXgw1MLxaLbW1tEV4E00lkDEGCTjSjdQnqAl+RZ/evlDwL6WpwlPRKtTQdiTH04Vgtr2/M8TpYuUE+VQEA2j6sUhH+Op0OJHFcZdI8lGdcYldaGgUVlVvfVgihA0IvUQV4kPgFcG/IqYA6Kwat2DJ2YEfu3bsHxq/Nuby87CDuXkCAFQOVbXx8nDxh29vbNiLcyW2vTji4nqQGuravk9AXG3Jmy3JFbPp2CPMKNv1YWNsnbP25K5q/WCx27ty5999/f319XeC+tY90LBPqZgPVej1Uqhx7HtiHYBzr5GQ9cfxwDgnm0iAjDfGvypxXoAf7qlar4QlgzckH64REOlQnalywUSHhEQ1BMn4gV0LJ9/b2+DIcDqOGd7vdZDKZyWS63S7FMM+ePbu9vU0GvomJiRs3bnzpS19CiGvxkDT1/PnzYE3eBBeO0+wxJl3veHD4wP/MXOCSSafTmggGPB6Pl0ql9957b35+vlaroUc7jkptwHA4PD4+HolEAL5nZmai0ejs7Oz6+nrluLEHOepg0IP+ERDAXCNl0KY5/tFgLIenDyzw0AjGPt9IsPbZGnIyaRc/thq+b1Ll6KyyEapMDarb6OgosUgQ8HVooalwtCtXF+cuU0l8mQiICpVQ8hP8bXopJc9AWPumKPFCnfePBJas+mfVSd8awFLOkSaqj1AqlRTo1UsfEVDoCPFeNDur/VkvaC84VT/RZyUnpAQUE2DzwT40aWQv/69Xj+i/Ur2/lYWB3Emn0xsbG0oCblOYyzF48qYoFdRYqSoW71IpVeVNVoZ7m7QI1Z71J7ooce1ahXJX8hMktU0h7/hCEA0sdJ47NjZWqVSINY/H45S54wJlTRKbhYjHXC6HqkG8cqlUeuKJJ65fv05ESSKReOWVVz796U//8R//seUDLCwsUJjRLi0RQyW+1fNHkhQOoG+rgYDvq3Yzfw0Gg+vr6zdu3Lh58yYGn2jOsmulspEettVqJZPJ8fFxpO3U1BRJ94+OjhDKIPuDg4Og1bIs8U+Cpdy5cwfskVnAi2u5dI5E9v6z197xdYFIUNpt20v6o8FIDXdi1geO20NjL8S2xkxRkS2tPT1LbwHRgEMOgJetgbPanvdsKNa/4gBED+NKJeSSf47IW7K9I999Dy3L+HBG7IH45zpexjoJe9XlAzYlIrxcLhPaj0agObBVTpw5dqStVeG954nX9vEWmbU6u/2TQqvtSWVz2vaR1/Z9vc/yWgkn8ZlYlwinbqvV2tzcnJub29raYnl5Y7TYaaq0YiupO0+0/8TmkmdZOXTA10RZsUWvLUam4RIo3O12tdScQq52KASqSrvneyeQzHqDFSOKjsx2ikajEK7hrpCtDRoPqM69e/dIB7qysjI8PEwt3UQice/ePbZZs9ksFovj4+NLS0uadDTxlZWVCxcuWF+QPIE2QvWhaKbDerSfCfBhHODqIbMsd75Wq7366qs3b97kTLK1krnJ8PBwNpuVkx/XYqfTmZ+fv3jxIvEvwvEDgQBh+vv7+5FIZHR0VIxpUiaVSiUYhILyVJNTk2Xn1Nmef6YmqVdeg6Q5PJOBh/3WcoIV1iuS3OHhIWkGNLzyP6Hq4iRPp9PiO6LutNttxKN6wlKRXqLFY535ogmwBjAi8aUrK6SvSmdjJoQa3Zf61ihWcQRft6msGHYLBks2my0Wi4R4ODVJfZlkynrl20Vv+RULyDilrawd4EwV+mA6nZZTW1ESWhAnWUNeo7iXZO+Db2rNeZMz4GHD88lhLq+6DFIOxU6ns729LV3D0k60jg8ODsjRTGoO0APY0PLgc6ZiFUHnoj+IY6kMdkhRcr2ljR2jh+st1umwR3SaoqQj/UkHQenFVqsF/48Ul7VabWpqigShjM/k5OTR0RGVyxuNxvb2djQaPX369NDQ0O7uLqt/Zmbm1q1baOWvvfba1772ta9//evUWOKwqdfre3t7qVQqmUwqQDkcDoOizMzMWBawpahq5DU15LER7AMhD00qkUjwjjiO9vf3k8lkIpGo1+sQnLe2tr75zW/Oz8/zdGVA5gAjaimZTJIhJJFInD17lghMLapCofDDH/7w8PCQg4Egz/n5eaae4teQuHZ2dnQG6/yORqN49TudDskMJCPYLFbw2cXsC0j6Qpd2hVgw197fyQysbDCW7Xp0bAGoGrLSFTgPVRQxKV/IcQjaINYvDDGLm3MrxlmscCkNUk3sO/Irm+iD1eKQmG0aS5JJSfQ7QIdVNHuZdK6fzfnghBJJh2K/qY4kb+U9+py7OS67gRM3C9v30ayd9aGJh6jruLnszU/YB+elel3g+xONm/fswbEAWcfJNAYPna0+MjICYQuqjDVE9Dq5XA4jTgyZWCyWTCbr9bpGDz4cZxjSn6ZyXGBiXuvBEocfz7PvO0TlcnlhYWF8fDydTqdSKTnB0HTIEAIQtLW1tbS0VC6XtXlQGLFOtKmchFY3btx47rnn3nrrLaGKkUiEDEegwDCXq9UqB0A2m8UlhWhAyzs6OrIVkQj7BhQWXnlwcEDQIGYKjiOcpeAw7XYbZBnP0A9+8IPbt2/jI3WSSwB9ItxjsdjY2FgkEqnX67dv31as0DPPPLOwsMDpThHLvb09OI71ej0ajebz+cuXL5dKpWq1KiHCCkSXFCBjZ1lr1aYAs8bWR9evfe9gqWJWo5fkih4HIiBtRegiaMP2k0NuZGQEDTqRSKjwtB1eyVnmiMXP8sbH4+VWOMmNHUnd/x0BrIV0eaFgXeZ8tlbOh+SC6GJWoHiT0nH3SCTC8gWstFRor6/W+/nkU+4IX9+hsc/1JgzCF+oFSbwD1P+vvYD4Xj23zGLvONhOSkzHYjG7SYaHh1OpFGoyIX/IXwuD2FTFtlpdIpFoHjfoYqAo1JFTdKWjIHvrFTnpoS0HwOv6H3iUpvt0u916vd7pdFKp1KlTpyhCKr0bJzb6UalUoo5Gp9PJ5XITExMzMzMQkxqNRqVSCQaDW1tb7777rg3iePvtt//z//w/f+utt9CSIFdR+LHRaGCFQCx5/fXXqSo9PT0tLUkKtdhK9FmmrhJR7e/vF4tFCYJOpwNtfH9/P5vNwtjb3d2lIMgPf/jDW7duQUWwBi+mD+kHcrlcKpWq1+vT09OFQgFUOp1O53K5ZDJ5586dXC73qU996p133sGqkEk0MTHx6U9/OpfL3b17l02Noq2Jk7AGIbRi0dY2tJQwiWxZZpYi5U2v4excu4UtEOoLJ0pAY2nh905+UGFLbnaHZ4kGiVmD6QZmiCDW/UulEucBKQ3sHvR6LEQO9o0X0QlxkrAacrpC4HnopvAVLA+oEfYib21DfS/jhfnmqeJve8NSvc5DZ/58m7dmnTX9Hmp82V9JW8TT4jziJGeGgzL3Ugr638QSVPint2CzkisJRmCQSeoNpgZVzlYiphq9jZAUUoFQxjeNWtdoNFSdRAtR9xGNz3FwWWbux6JYcRPtn2AwmEqlEMGS3e12m9wd6D7RaHRzcxNCVTKZPHfuXDKZrFQqb7zxRqVSOXXq1MzMDOQ/6ByKraf/L7/88mc+85l/+2//LUO9ubk5PDw8Pz/fbDbT6TQPevvtt6lyMDw8jIFiCVg2rQL3RCMD9hGCj7DD7wQLE2gLlQpixvDw8LVr15aWliSpbVoSsOnp6elcLlcoFJLJJKsCTkI+n6dXTC7gxhNPPPHuu++y2sk+FovF3n///StXriSTSXprM+4r76AtWSAsQs5AO1n9Cwo/XvONXoHCK84+lpOSjKeO3S1yhvGlFcQq1U2EUbPZBNywRabk9yOHhPVbaiLwaetLG/HgSCEbHN+/ATw6Gbq9gs5xFViB+eCXDuzrK/uUW12MWsYRHpWX16EOeTvnq53ZndBLKPsOgRxlEAfZAOSaYJ1h2ts6x0qN3auOl3Oo6PiVmmnxKTrsRH5ijnGk6YY2LRnWN14moqLz+TyyhmDFaDSaTqfhFwcCgc3NzU6nc+7cOZhbpF0H24Xu1m63S6USoJv1SyDly+UyGqv10/oeeycZ9pNQKvs0sewZuvX19UKhANkcJkMgEMjlcqOjoxsbG2SIDYfD2Wx2b29vcnKSOPK7d+8GAoHnn3++2+3+63/9rx1vj+5/cHCwsrLyzDPPZLPZ5eXlRCKxsbHxqU996vXXX186bghcJhTxd/78+Wq1qoy1TvZXMRPEcVR2UL4sFotK4qEaBblc7vDwcHt7e25ujrwliimT9KQ6wdmzZ8+fP49eWSqVQqFQoVBAypDakAO7UqmUy2VoDKDSlEZ78803V1ZWKMUAzuvgp1rGknSy8a3jC7HiW+7Euwbs1NstbCPXJUPoGGekVer5uU1OIPhl8IMqtGxn+y5wWoRxC2tWyImjxduU1jajjjfrXi8qhNUdvTpcHy8oaATmgg4JGzUquqGgeQEv948Wr2btldfWQYwcxEJHLqj4o7NjrX/ACjKnTJSjHffiVzgLTpOhv+JwBxwUpxUlFAMQ5NryE5zY8V4L0WG84ZK1m9Pr60A/whxjaQJByLxigukt8QggFYzztWvXxsbG+C3KEdrl2tpaNBpF147FYtVq9e7du0QSS/UjRMJmNZOuYQOCfV/ZK23/7DgAbL9YLLa6utpoNHZ3d1XvisYZQ15mioBgtxL68dZbb2Wz2RdeeGF1dfVb3/oWsUI6FK0jFCbAD37wg8997nO/8zu/QyROKpWStij3r2ydcrnMTMElcOpESznQbpcxqgy6QMnAO3CfV1dXm83m5uZmsVgkLxXCWnns2u32zMzMqVOnUJ9xFGN2wCWo1+sc8Jubm5zTm5ubKI9E4cdise9///tYk6pO4BU34kE69bqcMBYvotinOalvkIlSzqz6BV/eiZSWPadICNrW1pb27O5x9BDTyo7WSSBDzaIC1ovWK4e1r/RzVqnXsj/hb50GNRuHtg5pR9A7+d1sRc0PYda+r2EbfEyy8RINDIpEwIWqUfR5Z3Xa+d7qdA5/wLmyV/5rZe9TmsR8Pg/USwgAq1+QFuHXjo7pdMap1KDPbGAJfR34OoTI9SH0zVZTZhgR4sVikWUnSBozf3BwcGJiolQqIZGRNcFgcGpqCmotj6Y8eTgcLhaL09PT8I0oqwrRWFmqOa4FaIqs2Qd/12j0MoN8p9irwfVqGrpsNru/v080UD6ft+kH2GAYwrBceJ3R0dG33377Z37mZw4PD5eWlt58802bocX2XwyZvb297e3tz33ucxcuXFhaWqLqo1NQRvKCWr24CpF6GI5W6qnAK/XAlAwLhLRarZKLnMri9+7dg5COCwFEIhaLwdZQ+p2pqanx8XHVsSP1ErXHiHuEinft2rVMJkMewTt37rRarfPnzz/11FPz8/Ovv/46oXEDAwNkK7T8MKvG9poX7xRb/mIvsNG7g2Q78moOOM7/peEqTYW612w2YRnqAKYpR6DN7ChhreJqDkBBx5wEy73SwvRSFvt4yHppOc6dFaAPMVS5lXynw4nKfuB1P8mE8QPVBt3f38/n89lsNplMEgivH1oV26a/srCGlwbE/51ADF9hbQUBugMtkUhEIhHopVBWx8bGREfTgsODr6rJvCmxPBbRkwGl+2tjI4htgk2tVG/1P5uuT1X4lM8FXZuofdLxbG9vg+Hk8/lKpXLnzh2qnZISc2Bg4MKFCz/84Q8TiUSxWCQAfXx8HIW02+1Go9GJiQnU8FQqtbKyItxWYVTeiClveyQnqtcI873G+w27DmpatVpdWVlBM00kEtlsFpg+FotR0RwrAWuJyvfRaPT111/f39/f3d1FEe61YSDqZTKZ7373uz/3cz/H9bVazR6uth489BIgaao2Ky+rbssakFGlJxJBvr29Xa/Xl5eXSWBC8V+lDAwGg2IF6OQeGhrCezw0NJQ7bvF4HJspFAoR0oVplc1mwYWwnxKJxLlz5+bm5tbX1yORyMrKCvQSMG6mnr5ZNoujstkdaifLqY/caw1wH7sfkZuIC4lgTk1VfLd+Nptbg6r2Nh2jIgODx/xCRxvgS42wYxn0Opm86PAJW/+IPPuNcwGHFi+Ckc3s+Ma7saPtN/dtZQdzROhgoMHStT+o1+tDQ0NPPvlkKpW6T0I4un8LVaNwLCb7AopjtjkHELg4eVjlwisymQyDwoaxI6XPlUpFGm4mk2GC+S1/BX4BpUEiE1LhiBh7kEhAI00klJUpn6XGEseZo/tYTc0KRFUDEUMe2c02AHGGTwYblGVHCp7d3d1cLkdpGLJwpFIpeQIpGgD3FoAVMtnGxkYqlSJOD8sL25+xchZBrwUnbdG7Ub0HJ12S68Kp6mBHQ4lHiDVPJpNTU1Pr6+tbW1sk2Dt16hTpLC5evDg5OVmv12G/gNIeHBxsbW0lk8m7d+9ubGxw9BYKBRanch7BytA8gvIvLS199atfpQ9LS0s/93M/9/bbb3OwodlRjpYMq2RpgC3ABlNFWnEJlDsiHo8TflWpVOr1eqlU2tra4jAGuxAqLZyKBLMkYEskEqOjo0RdXrp06eDggKAMMofk83mwr729PSI2c7ncj3/847W1tUgk8txzz926dat+3IrFIh3OZrPoJRzPhEorOQxMdmYqGo1So0dRjlrzNq0xhg4ThyWBIcv9tbS0R4Dd2HQ2CFM35KSRH3hvby+ZTBJMoOmzAaXDH4RZ6eSWKBAA4k2/Y2uEOn+yC/uh/nMvSnwSj5qjxNhqGOAQxGfZbltfDm+ElkC8yE9zpClqSxn6Dw8Ps9msbhSNRrFGWUkqxIszxOmf1VL1/iwCq7FyqGYyGVBm1gdONhwFOmMdEEouCM7qRqMBVHd4eLi1taXOaIfoh/bslaFk+6n/Q4lTKBTrTzU9WTeWueHACF7TQV2CSxcIBDD3IBQ/9dRTCg0FXk8mk0tLS5lMJhqNxmKxzc3Ndru9srLCiCHN8Uyi62lzEg3IQYuiLdVGepa34Gl/vdirPlsmiU2zyfrzxmEqcoxHl0ql9fV1ElsjTKkPIOaW8tVw/ADj7O3tQakG8HG0EgATaws77cUXX/xP/9P/9F/9q3+1trb2i7/4i2+88UY0GsVtLq9OoVAA5cBupXBtpVJRFSFVC2QNwKQGhla8CYlTcAJjRfkOLFlrJicnOVMzmcw777wzPT3daDQwuQKBQKVSabVauVwOCGVoaGhubm53d7dQKGSz2Rs3bvAKjUYDiDyTyczOzgKRAfGziZRyyNafg4OrFEVSikUnJSmrzbCBc4WFREAQSaNs+KsWhtKuennHzmfLqHkMJqhtH/0O3taLxmYhXCfHt+NoZTwx9GEfxGIx1SnlbgcHB5zNKgfBlZyCI9PT01YvFnZBYjCrLaJoo2MWi0XWDbVWfRm7znuKVWY1Mg5MUDm8bWwYgFfWhNdLa9eBQ3JE09HFtlyQRtbmdfSlgnopRIwdWgOSVIPrvKadQu98K18BUcKpVGpmZoYi3wsLC0888USlUlF8fLFYJBIPX9Pc3BxVrzqdDto65nCr1cJ/MDk5ubGxgX6NTg3bhPMPGx/J4ps75qFmoxetk0wUAxfihI1i9+o7iIydnZ1isTg1NVUsFpUYIJfLUccApJhkF91ud2trC2lObkgqV2EqWcYVS1Rr2Jvu6vbt23/+z//5RCJRqVQUUBOJRGAHIX9xBlBaiGVJqRebZ1yOR2nKYA5CuvB140xT5LR3qbTb7dnZ2XK5PDU1RWbB2dnZfD4P/EVsNI7H7e1tAhpXV1fb7TaV3XF7HBwcbG9vQ1/BmZHL5eiPDmyYJIw8iUTUEwwXJ/2pqvqy5RlMwilYcoge0SEcx4aXMvBQJ54KM9nKan3cXQN9dWHnGmfYH6P58qa85eusF9dybOQrEhjAsEt+cmcyc6F0o6QC/gBp3N9OwnYREyRX3d7etmXWUIVsEmeC5dEOrNIq0WzTHvKNgv31JUsKQ1IuiE6ns76+bqtMeofb0a+tRuBA5zKULM/fawQ5TVx3GrqJkrZYx4vNOOrrdrD5FoDqkskkyRih94Mwzs/P37t3D4kQj8fL5fLdu3dXV1cRBPF4vF6vQ6kmwIGQNtkTe3t7sViMKsOMKqKQE5cAMOzfra0tOwuO9ePQcp3xsQHutthKOp3GU4qQ0qIUoUqDKY2yWq0mk0k0i2q1il8xf9yI26zVaru7u8TIkByuVCrhg1WiBo2tTk25agl1cQjCe3t7t2/fvnr16uuvv16tVsl2rWgX5crgFJShUy6XxSSzphjzjvosCxcppszFXh3fprthH05PTxMXvru7yzapVCpTU1OxWGxvb+/UqVOUgwEwbDab4XCYtNcTExOVSmV9fX13dxe1JhgMkmY2l8tVq9VUKiWLEGNFwJ3lBVlOgghLhJLZxUxDd2FZMtosJ8mBxxCL7HoFczpZmv8s2tHD0OoTel+8QslJFighoFJKZJEMBALQB6zbgxRA+iYcDiut7n37b3l52ekEceuUQcJpDinS5lq1hAHLZLSC1Sl95jVPtMe2t7c5JNgeJJ626puTy8nB2a0lZcfL91cOV09i1JkGQFVGE8MW4JhINpkI9q0lxez7ckhaK0nFPhBPqn03PT39zjvvTE5OMkMEiON8Ozg42Nzc1J3JZ4Q/s9VqYYoeHh7eu3ePuQdVIGwaOzedTiPBx8bGRkdHV1ZWoOhKe0KwKhuc0t/QVTlFw+EwexhR6FAhOUhgIuImskNqecqNRuPdd98dGBhYWlpCjlSr1Xw+H4vFyuVyPB7PZDLVanVzcxNfKy47TggI/qKQosujve7t7eEbZAAhJExOTnLIAWG99tprf+2v/bUf/ehHnU7n8uXLeD4gOKVSKYQyyiP4QKvV4rC06X+pfgCSK0QYMcqAiHRko2ets4ScAWNjY7VabXl5OZPJ/NEf/VEulzt79mw+nydcc3BwkOMWPSkcDi8uLuZyuW9961vdbpeR2djY2NzctOm6Tp06lcvlBgcH8X8IbhYih/9DsCd1dixWJn3ZZgSSsNayV0URHWBaDw/Fc7XpdNySTQEqiPI7qqD7wIdRZof50/9Z3kd7rW0vWKrtxgA6ji5vTUttf5njAJLaJoww6xBFFlgJ8xfdVOmf0OqkGhLg8wBNs4cDe8DyvbE3bZyPQ7Nz2LtSe/sPpb2D8GU0qYf+sM8FXjjCMVKc9/VKansUsToBZJzUhfqhVVXs0SJ9n2MJaVKr1eLx+Pr6eiaTgfiF5kiE4VtvvTU5OUkMC5UcSJMPJqDbZjKZcrnMpBCgRdUr0mikUql8Po+WPTg4iNOJDP2wvoAvOed1aiIOWFupVEqp00WYZXvHYjFWiKhIuNHUNxVX9JaD0mqpVqskDgavV+paKpWA2qPeQqTDhpCkUCad7e1t5SPG4Ein0wj0Wq1WLpfJxXH27FlS7MfjcZTuSCTyzjvv/Lk/9+euX78uWIzRpswKpR6AsycmJlBUNaeKaZQoFBlfNdgc20XN/rNUKvETUrjk83mCA8bHxxHTg4ODi4uLqNW7u7uJROKHP/zh6Ohop9PZ3d0F+iA2TYUsqJkrJpKWvarlHR0dbW9vc05z5in3rBXrjnDU8lalMZuux7txHkkpVs6sP1NtevDDYkSHgcOH0QKTCqiaL1ZWWLJyLBZzTBNbF1jJXtQB9g6bCOUY8VKpVDAyoDw5c3Ff2/K+EiqJqKPj4+OlUsnazo5wdNJM98KqfAkutv48x74FwfUIDYQv1NCL22uRL3tseoF1p7eOWBeAYzPqSnBwjQJnbOeVPIhbQbN79913L126RDXYYrG4tbVF8n7qWiETOVrJDgG0CufB+vfQRLB29/b2bt68CReCmYZgA1kYc5hpRSiHQ+FA8L7qBxSjepjeyEyNFYaq0sJB7KHnKudmjytf1wUfgHQCgcDGxgYqFdoEHQYZA5dHx2Rp4XBGmmuriOpDTlQUcPLzkQ8SbsbY2Bjqdjwe//GPf/xLv/RLf/AHfwBjjOMHaZXJZKheSAIggpX5oYUEw+EwihLBB2J5S7Fy8j/YgmEOZo1Zg6eBGJCZmZl2u/3+++8/+eST3W43lUrNzc1RXeG73/0uQk016fkh275cLo+OjiLZOX1xe9Kcnc/79lIyvNaAo1l7g8WtGe3gJ94X7yVMvK2PABk6Ma/f/sQx9527yRMm3YKqPVgh1lhMJBICMZwDxpbBslWx7LNarZb4ozJqoQmQ9lrVZySTfypfdCMcaPQjl8vFYrHFxcVmsylveC+PnDPHj9R4McxYUVssAOT9yUNdFl6d2tfSeagDRAQSm+bRseOc6/WBz2RAh7xB0e6dnZ3FxcXz58/Pzc1lj9vCwkK5XC6VSoVCgewE8eMm+qr2P2gDjElwTB0nIK0qV6Gk9TB58vk8Yff3def0/dJWYsICEwsGQfzhRWAliQlDmVrmaG9vj0SmTlF261ex9H7tdjiLUAjQXsmwgezDEJbzUPw5clGxuNFeEamQ5Cw81W63b926FYlECoXC888/Pzc3t7i4SFRRuVx+5513/sbf+Bvf+ta3MEgVO4pf/ejoKJlMKnAfeFG6EhO6s7MDjTIcDqPHKCbNN72G0+Saw2ORSCQ4kjmbGfDTp0/fuXOnUqk0m83p6emhoaHvfe97+Xx+bW0Nfjp7Hj+T0qATiM/CQ+N2+Jf0n6Awmq196kt0s5RkEYocA8LXH3iSHdrrG2+CIKf5qgK+B4P9AEPfOpMUFGLrTFq5oZy3MH+sAseYO0JcZjQsNaXPtT4zmwPWRjPiWFKeAPlCHtjoVnLpHWTwPv/88yRPEK/A0Ryd17OMgpNo1oxXMBgkYQ08Njvx/KRXrn3vnW1BGV3vZebZX/kuKS/ryNJxBAvY4ls2OMjOCkclUum5554D1X3mmWfW1tauXbuGSRsMBskJPj4+DpzNPOGL4HS1wU6wrRUTIZF648YN0RUIiiEH0Pr6Oj5AMA2bW51EE8RAV6tVOoz4A0221h9aG5JdeUrxgzkjbOvRacTwnu3v78NpkcMql8tRwpyEcyQs3d7ePjo6mp2dRZ0HRGJjIEOt+cn3kjgsdNwha2trly9fPn369O3bt5988knqyZLLe2lp6ROf+MTa2lqxWMxkMqI8QoATOIDiD3jNnbFpkNS8LGlsMVwODw/JYV2tVolhGR0d3dzcZDZBbHQE4g7d3NykVvLm5uYnPvGJQCCwuroK0NFsNrPZ7NraWiKRqNVqSs3aaDQQ4u12u1gszs/P53I5YsG0ubwFQER4lU9VUYIW+nP2iJ1Kux8RMbIvnbpZ/fesV9rKXY+v0nHdH/k58MVfFjImAqVVvZPJpMrUMkcEA3OYOc0XG0B3Ye7IsaNOSqdh1xDYYaUBzXrUnHwgVvaidQFYS5ByoH6IXi48DpB6b2/vm9/8Js9WrZ2PF1RCNBAPEolEqtUq/i47cPYA8M63uu186ajPvrRiL5zifOO9wHsweDtjUReuAUbEQh8dHZVmSiADAqvRaBQKBULmGBBmTpF1TIEYF5FIBLEuzHd4ePjSpUuVSuXNN9/c3d2t1+tkAiIb1FNPPQXmUCgUwMElfYhRBvUmKloJ2JS3V8eSOA8AI/zJCnRvmQXvEKnoEQnkAH/IgoKnDstAaLgG0zHdONotP8d3jV2/fj2dTp87d65cLj/11FPvv//+66+//sILL/zgBz/49V//9ffeew/SKwcYeXYQ2ZVKJRaL5fP59fV1hovRJquJ0A9BHCTYI89tLBabn5+Xu4ntWigU1tbWbD5S9RMdH8y9UCi8+uqrRI1fuHDh9u3boPZEusJPx/ELvRrof3Z2Vm5Gr8fbCxg66KWzNRwx7dU6LSbr4Ga95sJ5hK8kUTQ2wJcttXH04fImVmdCxuF0AVnKZDJKD2sTlUi4w3BVH2wKWS14nHvqG5qTyj3r8K7Vajabrt6uV2ZB5zi0Bo0vSvMghMIr6bAKbbUI0JNeRk2vs6iXg9FZLnAMeFXtYSdLp8No9jaHiu8Mja+PsRe27iyg/qUSHorIK+onHo8TF4dZDWF2d3f37t27ABrJZDIej9ukCgjxWCyGxgcLQlVUVM9CthiTVa/Xr169Oj8/v7q6SkAE2Mj29jYBI+x5ZgcFkLwci4uLWLWsSKvUCHa3xACpyU7WBd9h1Jc00Ui5w87ODpggsd3NZhM+3N7eHkXFbJFZJyWmgnF6zRGUm2azefv27dnZWfKuvPXWW3/7b//tb3zjG2BQlPemz41Gg5DabrebTqcTiUTpuFmSaLfbJWhWxRwwmKA2c8QmEgnEMSIeC3d0dJTkqOLAqZ+8Iybs7/7u76K4/NIv/VKlUolGo5OTk7u7u81ms1arzczMwHfEQKlWq3C6SM+vFxGZ1ZkRC0xJ8Fn6gLOYHXzZ0VoEZNnd3UtYW0TC6/fSbdErWbrieh9+4CJyiniwclCcxcYRs5DD1cKYQvmU+5QbwtuxheTpko24tmaEOOycE4oqUNS788qqJNnnaLRD4eC0971ijrixGUc107aM/AmbXtgrvm0P5Moncho7AliA+zhv7pvIyfvoXn/qjyQ6+oijlXuv9AVVHKX+8PCQIOlU6j5GPDo6SsXSd999l5MJNGBqaiqXy5ELAtWA1S+tWTl2EXOkEmW4ZIvpqEMHhwWo5E03b97EE7W+vn716lWiLTDnVcMCPR2xaDVZB/FnQBR2Dxn5oUvCrkiegrCm9CJEoLNnz+KuJOUeFDfUCqnwuqHel0HrZfBhLpBk4+7du8lk8ud+7ue+9a1vBQKB0dHRnZ2dy5cv3717V3HkBBmhnfH6ZHBtNBrKcIujWLxgTiwOQpiRoIjUHCBhXjqdRptDlMPgtNoDhd5J/HTv3r3R0dEnnniC8giEug0ODm5uboZCoWKxiIcTeX1wcJBIJE6fPs0aY51g71tQUYwdJ7lYn/Lh1sNsEV7vZYDmj+Tx87q79Cf0FfJ8MXEjx6aJrwwBCtNfFSEhZqFkl/znjeMmGr7XLrffqGCQI3aVw1JZZMW087ZHgu+98AhI4IccjGwnFf/W2yoWwPcBvshv/x7ouCY2T9lY5Duyae+dX/m+jIOS+yr70MKcWGSVY5dqIKnEWc1piWlsR4OnyPvn4FBMcK1WK5VK9+7dO3fuHNhlPB4Hc1BUHv7Gn//5n2dFkqdN64A6IwhizDG2N6iZFpaDxvB9MBgcGxsj2K9cLkOlKJfLMzMzc3Nz4+PjBLLrJBgbG+PD9PT04uIiSbQFHTrwIv5A1bq055wWgAOMAGt0Oh0KBPNz1fyEqLe0tHTlyhXU20qlIujclgZVs/qERV2dBSmGOOToO3fu5PP5iYmJW7duXbp06cc//vH58+dZ7UTHZLNZEHA2gvLY7e7uwsve3t4+e/bs8vIy0Fa73R4dHd3b2ysWi0QqHRwcTE1NJRIJqrpUKhWSTTPCkKDD4TCFDiC3QDm4cOFCoVD4oz/6o6Ojo0KhMDk5uby8TDgiHDssD2lU6XQ6m83u7OxQ0wB3VrFY9CU7j42NwQkTCVLBh069Ea9/3ndgfdN78kbynmmna1PLK8aeokEnVfogQEIsFT6MfJAc1ZL66YkqwogRxAGGlaPEjQL0YMp6cx/5Wsm2PK6+12fviOlX3kQO9jixks0ZcKeaAZcB0XzobPflDvfK5tO/SbZ6mde6PyebUtyyWxKJBDwBx/Ky4lh2je7vxdGsOmABL6sh2n5yK0ENBIOIC0kRdyVIQTNyeqh+UqyzVCrhOvvxj3/cbDZ/8pOfUH+kWq2Wy+ULFy5Uq9XBwcHR0dHnn3+efDrkh8Nmxyizzgp5q2WpyShzzChrnAaDQYowpNNp5fDc2toiJ6cIGJw6REKdOnUKEj4dcExp9UfVkvp77e26R5UGVXCUGm6ytbUF8pBMJqkDIEqTk3/ZvqzlzHo7IHWPzdloNO7du/fCCy9897vf/fVf//Xvfe97586de/vttznLoYugyAiR63a7qNs7Ozv37t27cOHC1tYW6gU5XVk2uAohj+OqzWaz0DNYTrgTYRcAQJFga2dnp9Pp/IW/8Bdyudy//Jf/khJWV69efemll2ZnZ1lIIFoEN+VyubW1tY2NDbyyi4uLs7OzFv3wnYiNjQ0EH2U8sWYQiAMfR7PRNEozII4gejf7ms/kFZB5YSmhYiJpQg+Ow+XlxLNUUYAyoRMaBCUbIQSfB6EuMETOwn4k+da/PXZaEqcIjnMYPMCseUP2nqWIOVCgL+bbp/mePALsSC0PD0QVpqnuammq1svnHHfeGEXZ745a7bi8nHSsQmOSyWQ4HI7H4yxikS6h3NqwTCFQDgtV+BfU3WAw+IUvfKHVauXzeWBTMjPcuHGDpH2k0l9aWoKCRk/wKcu0VECwNyu8nSlnKOwhpHgNItZYT8rSlzpugUCAzKsUMQE8xdzztZTVw/4mni2ESleXl5c5qLwX7+/vr62tTUxMQK6gfcTgYxEuEf2hUGhlZeUnP/nJxYsXI5HI+vq6zXmE/YH/Fj8htAEyIyKRyQGgAmzhcJjkiKh1RAxR55dZCwQC2Wx2e3ubDE3JZBJCC8gstvP58+cvXLjw3e9+t1arjY+Pf/WrX/3e9753+vRpiitKAuLMuHbtGvXmCSNKp9NjY2PYN3Y1WtaXdFveTvVAvEJKK+1R604o94VT7Io1hq+FPAEy5mzFZ6lQZAG0OaFUIHD/g4KKTgETvQIhGhJ2EPZhSQOksMtUodA3/MIrc/oQGbzf+zbtViezm9N0mDm/fWCdOHX2vFV8rLruJTw4V2r4nOTZXgMKcUC2dYxcqvDJLLLYgvoAp9i+s4XShO+oh6rjZxEDJwexjQSBkQOtslKpwE/Y2toCoOAkQ47gKrTvLmcFMhrnPq6MU6dO8YjDw8NXX30VoywQCMzMzIyMjGxtbSmPinRJRUyB39kyKA49UR3ohUQdHh7m8/larRYKhWZnZ6F/7e/v3717l+keGxubmZnJ5XLZbBYNVDRPUiTL429nHD1RKoDjFXFMOSadijAcY9Ju7ALDb4E5b+tJPqqkdhaq1gDgbzgcXlpa+tznPre5uTk9Pb26usrL4r1EhqLewsxDs04mk4DR5XKZTO6EyBOyJIFIXAO1MZk1NpRKIXMfrXao61/96lfv3bs3Pz9fLBb/8l/+y//P//P/kKIHFIV0MaFQaG1tjaAHcjaxbAqFQiaTUTZgVQhyDksEnBAVkiCitjvuRPu5F57r5NTEPtbmkprPn+A4sXcgrli9G3eFWEZynjsV2gbMqnY0Eq1/5AZHYCwWU/9t6hgmVylY+6wf+9x/Pw0KgLe2OF29v1WwDnQqMqkyWxh3vMxa+vi+2WnCmEhkoWssXqkiQ8ojlc/nk8kkWiSyDwiVBNmWFqNOO/HiYBH4fxGROzs7BHoAWon9hrqEw8r38EB3ltGEYK1Wq2Q1w/+p/P3EamupOS5QjfjY2Fg+n2+1Wjs7O5VKJZ1Op1Kp5eXleDz+p3/6p6OjoxMTEzs7O1KILOLkwK9O+kCvM80BneyXml+0eDIvk+eEiEqkMMgPqQweZI05llaDg4Nnz54lAQVCXLWxYQ3RKyc8T6vckro4q6Cg2d/aYwY1c2NjA2KccBsFDrTb7UKhIKHANLH8LOSFhlsoFMRlTqVS5FEplUqshLfeeuvSpUvnz59/5513Ll68iO+RMRwbG4NnzbDH4/HV1VVqaCWTSfyKyWSSpCvw1svlcjabxUsBUYT+U28MK4rRo2IR2AiMkV//9V8/PDz89re/PTIy8pWvfGVubm5paYnwnP39/ampqUAgsLCwgNQ+Ojpi2aAzTk9PX7lyBYuBQ8LhgWlSLPWWiHzUXpVmV2gPw8jBJmEhzA1HDgmJ2KcoNzaBEenD5ACnZKhVpBApWjPKta1vfNO3+TZfB5UFvmUeSQo5BQrsfZw796I29sE6HuowdKBmew2blH9aPB3r9oH6Bg/RYqBMz9jYGOOby+VUaQVUF70gmUwixdjkaCLe8cWLSEPos7sotqsUE9xQkQ6Wsei9LWxllVYSPZY9rwzXsjT7pBxhW4pZwTbTYNmydSrZ5QX0nRjrnZ2dwcHBTCYzMzPDpm00Guz80dFRJXORF8sqLL5Zfa2r3bculwNb207yJw5j8FNsUrgl0EvIN8JOg/CrLN7k/COnMwqUZZ45c80+kXnL06Xx9UqjaLUkSN8adugN5AvNZrO5XO7555/PZDJjY2O3bt1qNBosg1qtRggvEITosbxRsVgk6ioajWaz2XK5vLS09NRTT01NTb3xxhsTExM4dSFWEviABjo0NJTP51dXV9EkGBAwU3IH1mq1sbGxYrFI+AwQtiBvXqTVahGmGA6HK5UKUBvB7k888cSXv/zl/+a/+W8mJyenpqbGxsa++c1vXr16tVarRaPRmZmZ27dvr62tEXnENjx16hQxMru7u6Ojo76JonS6a6itR9om4kAW28xcVlFFtbJgAjlngKcikYjwaCYCoQ8Din/agKle6qpdBn1Ywn2aV6SKUu0k9dfxM/AfqPWX8rYEpXXO3zfRZmdndSmqqKoLs9MIfECa6y5UuGA1A96BweHpdqAVp/Fs8EGVLoSpo7wKFhHGs+eV2jiCAoFAJpMJBAL4GUi6Rt0m3BTqhq1R5IyaCLwcABxuioiTGLWDqG7YD7qgXq/n83nEGencSqXS3Nzc9vY2BDUOeQUKy6HqsFwcNM1a9I6B79Uv7OltazgpuAZLiNxS5HHGP5bP5wnPUTAVtgLJ55SXg/z64s9Jo1GNWqtZW0zjJKalzqTx8fHp6elTx216ejoajWYymampKeEA5CcA0iWRITpsq9VaWlqiYkuhUFg5bq1Wi+y7LNq5uTkKEx8cHAA6q2A8zmTeCBgatBeVGVYP3A8uIyYezIrqxrwsv0UBh/GytrZWr9cpfJNOp//6X//rf//v//3R0dFCofD5z3/+H/2jf3R0dERtgVwu9+KLLwKpg9iQKSWfz09PT29tbTEISFLfHJMOH06frV9RbBPrJBSyQd4YRDPrDbaSasnjUMXq4j7oAXq6ADRNvRd8sLizl13mfRfbHD+/vkeAwIJXiK/N9P/YzXsHX5ng/MT3TPW90quW8f/7nmXFpzsFmyEDEdlMaKYS9bLJA4EA6iHnIRarOmHNHKsJ4r9ScQ1bhEX1PeURRl44zDC5Q8H7IJNQD4GAXcEgen8rpvvMPatfebEtR8fri3B4DjY8vVwu37hxo1gsnjt3bnx8fHt7u9PpzM7Ojo6O/uQnP0HFzuVyImNIm7DSH3hBa9erU5+kOJ73RBERe3h4+PTp05VKhQNjbW0NnKFSqTQajVOnTgGCEVkKhKKDXJALx63NaW4zc2ncwKNkhQycoF24cOHnfu7nKKCzvb29u7v7ve99r16vM5hYTolEAuNMYUTSHMPh8MzMzJe+9KVnn312f3//i1/8IlE2q6ura2tr169fX15e5tQ5f/784uLizMxMsViE7cPBz3KKxWJo5TqZENYABbu7u5FIhFvF43FVKYrH4zhglRQbS1wVuyFNfuUrX/nRj360vr4+Pj7+wgsv/G//2/+G7jI6OppOp1966SXKugcCgUuXLiF08vn8xsYGdgBlOlixjrD2Nlsj0TrYM5mMhUGEJgs0oP8SylbqsccFj7AlMUosjufsF1vL3Lc9km9zsPd9EC+i/cmq+xiJH97O9NJQHSPSF37xxpFYgHSENDQWdqAxZ/B7YIyCewaDQSqBAkiJ3iTj3TKOxZ5xevAYr+3NFo1H/ujoCCrF3t4ecgePsJeXY6vsWKNb97dpLWUB2J54z23nFCHMBC0J5at23M6dO7e9vQ2aiSGZzWY52BBzjmUkx6nGU//0YgiWj+FNZua47yTLlPgik8m0Wq1Go4FFDwQJVDU9PZ3NZsmIDzxF53EP5HK5crms8tUc5KplwedQKAQlsdvtFovFpaWlxcVF4hEE/ig1LgYcnucvf/nLn/rUp9bW1n7/939/fn5eb+cQ7eUjkT0kbYPQvvfee4/ETJgCU1NTly9fPnXq1K/8yq9ks1nEzdNPP/3OO+9Qg4aFJH8yJqCq9LIjqMszPDy8s7MzMjICrCG5psqfxM2D/pNQBVUgnU632+3Tp0+THuAf/aN/FAwGf/EXf/GVV14B7pidnV1dXX3nnXf4Ffj+xsYG8p0CeCMjI1RvUDyExgTTh/+jdMMxt6sXCimZxWw4te5DGlvsVExAFdi0Gc28m1q2vO2S81ff5qxbW2i7PyXRt9nMU8ohqrw3vYhqvW710L/qfSUP+aev6GOd2yxOllmAImUBACDZ+2bf3NycbmfJfVZ8tFqtTqfDyqAg0+bmptWCHePFVkmwYrGXgWNbL8luB9cBAVSwEZDa4bR5J8YR+ifRTB2F1/slG5WsidRvJMQcQ/jb3/42F8zNzU1MTKRSqe3tbWVfs0tfeqjveeurj9hENr1oIb7njRYZ+EYsFqNS1O7u7t7eHonZ2KvYLtYHhfdpcnIS76W3ao9gEKLytre3i8UiNQTIfGLtBpYjqyufz//mb/5mqVT6J//kn4B+4vdz2OXeObLr2wbyYCCSfaXVat28eRPN48KFC9Fo9OLFi5cuXfrc5z5HjprBwcFSqbS5ubmyslKpVNbW1paXl8mUreityclJqDVYb/QNLO7s2bOLi4vyBCiyhouJpeSsKpfLf/Wv/tXf/u3fzmQy+Xz+8PDw3/27fzc8PPypT32qXq/fuHFDoUbKS84SshmgstksvHjqDPDuQC69gAX+HwqFqM1EEUhdybIUWKFFjv/GyS9qNWXL+em19pzv+2+6Ptcf9ZbdXreNQnJsCNWfXbMJF50ajLJmbI4zSVG6p4MKaiMaDyDHg9wgvUBYKY+wKXC2YOOzrwCXvcVAvaKN5v3s64Ht46V1lp2V114Z7Xz2invv/Xs154ZePEQHbO64UeyGtMjUPXn11VeDweDU1BQ4NejHyYE539fRZzlSfI8039f37mdVnMGdUKvVVlZWsKsoZYBulUwmkc62hixrlNWCgiA3AAVn4QUTFoT1gwZgE7EODg4mk8m/9tf+2rVr1+7evQtDSamjLFXAQSd9x0dGBksUqpzNdhuPx0+fPn3x4sXp6WncBipQeerUqUuXLqH1r6+vLy0traysXLt2rVgsAowQnMIei8fj+/v7uVyOYu0WcNA1YjtQcjqfz09OTt69e5fa7V/5ylf+7t/9u8Fg8Jd/+ZeXl5fffPNNHJWtVmt3dxcNHbdwOp0GJqa6ENw+mPKSkpyySoVsfTYaH16kWCxi3slos7vPSXtrPeoWV7TypU+iNOmPzl99m68n5qg3bcN3d9huiwnqKHOP13xlmm80tV2N1mdoCRd4B6HnWvEi6pccvx9Kv6KANF+pIaYaCoJMeEuV9YowmTNenc55RC/Oo6+Q9WICjlLvdMM+/aHoXv/mFQd8yelVKBSIIEePwy7JZrNPPPHE8vIy6aqhWzCGWvcW+O6jkni7QWMbWJ5/r8qQVv2xVC2EyOnTp6vVKq65lZWVbrc7Pj5+9uxZjHf8UaQcWVlZOXPmDGgY/eccQp3k+5s3bzYajU9+8pPhcPj8+fNPPfUUBYIzmczS0tLNmzcXFxdv374NJD04OPi3//bffuONN77//e/jpCKu1RLnTzIvzj/5AABFzDeytVqt3rp1K5FITE5OPvPMM/oh8YRbW1urq6uVSoWo2l/7tV/7K3/lrxDMvb29TSHjubk5kFw4o5BY0G2hNil9AvgSuBA40nPPPfc7v/M7kUjkN3/zN3/7t387m81+8Ytf3NjYuHbtWiwWI6KVLH3KTg5ubrNZpdNpKknaFQ4rAysTmYtx4wRwqUaJY7o5JfqsRHYULNlDXt+Jrw3tJPA7Cf5wcuR6oMc9ObekaFpU1rcDj/TEXoCqPqPNOEJcDmerEyt5g1JdowAx0eVy+b5mzWyBUdruKtAW8xAY5fDwEABR5JD+5bv0DjBbrSR1InkcM9bOvXW+2fmgP0B4RCUo23Kf/Jz6vxBn62xxLnYK/9i/OhnK+efIyMjCwgJcYLZZpVJZXFxcXV3tdrtnzpxJpVITExM4PbCGxKy05X11znlzoXgXhNd/aK0NgWKWl6lcUTrheYVEIkFuSYjYg4ODzWazUqmQbmJ8fBw+H2QJ8r6isW5sbOBq40RHPN28ebNYLO7s7IRCoQsXLuTz+VOnTq2srGQymTNnzjQajYsXLz733HMkIL1582YkEjl37tw//If/UEn+oPP7zo6dEa/r1R5RxEeAVwApxONxfFyJRIITBS2bnJkUW0kmk6Ojo7du3XrrrbdWVlb+6T/9p6Ojo+DdP/MzP/PMM88g7Mj70W63N49bKBQaGxtLp9PUzcIMxaO4v79P/F63233yySd/+MMfBgKBz3/+8z/4wQ9yudzMzMzi4uLGxkY2m4WkTPqX8+fP01vogLu7u9FoFE8v6F86nabuMI4HnqIMNqDVwB3ZbJY/sZfhGmoYLZRhx9l3X2tFYdIJqJWbx6u0WQePs277HMO9UhDb1ivFq70AXUqcFlVvUXUkaTbYeU66VOFp0JxAFFR6SQJHJCglX/N6jGhCyZy0S81mE41HT+cD6MUDyo5jY+pE5TxX8lxCch0jpU/rlc2k15j63sTXkNdP0FZk4LA0nd9614c3kv4xmhPG0u127969e+rUKfRcJSEaHh4+c+bMtWvXarXa1NQUIlgRtE7lTWGytryWM3Q2rs+mB5CC41Q4JRCDqRRrAiHOIWrHij5fuHAhnU5Tt7DZbM7Pz1OCHXMK1y4ZsQFziXvGjKDGDbSN7e1tKiU+8cQTxWKR5MuhUGh9ff3VV189PDx8//33yfkwMzNz4cKFr3/966jStmN2wPtoMd4mhUDVlVB4xegYGxubnp6enJykgKykyfDwcCqV4ugqFourq6uNRuPu3buvvfYa8HQ6nb5y5cqzzz47OTk5MzMzNjZGIC6HXCQSgbKyt7e3u7u7ftzI41GtVoE4lpeXYXDeuXMHV+rCwgKyHvWIcbh+/TrbsN1u0yuGnewiAwMDp06dmp+fB2oj6BGnIognGWZIiqLqX1QF87VXnP1o1RHvBrTGn12cvht/4M+sDfaV1EJppCnaQC0lKpGCbLckTYH+uNCdeBQ2srazBLfMGgXlWYNe6cmcfe1QJ/U9s/ZAWFsoSp1QTjVyd0QikY2NDRCMXllDvQPnG1PXy1vqIE29ht7eHOVOx5Q37tnpm8DlXkSZh62ND72LbZFIBO/c4OBgpVI5c+bM4OAg1U8ikQgBEQRtKgm1zYqgbig2z8kSoLSWTpV7Be/g99Mw2mUXj8e1kuT3cIBgov4UysHdYIM0Go1isQgRKBKJ5HI5+F6tVgvMNHLcdGcC9C9fvkyoOi67+fn5l19+OZVKkYuOBBrEDa6vrycSiT/90z9dWFhAibBlcRyCqjVr+ufR5holmlCeTFjtcE+z2ezExEQ2m2U/OHSaQqFw5cqVa9euseDRRuv1+r17927fvv3666+PjY2dOXMmGAyeO3fu4sWL6ONkvDk6PBoavq9/oS6VSqW1tbWdnR2cAT/zMz9z8eLF73znOxcvXqzX65ubm9lsdmZmJhaL1Wq1e/fuEa8IBoL2SqY6attLiFDamLDPcDiMnxP9HQqdtEXcJIwD6LwvWOHgD4/h4HFEti8Q99gg5OM18XHFMVeCMGnEkJLlGNdvyTJvRbx0cMEXRAVyjXw27C8OXXzU3NCSCL0Wv2/n+f6B3q6sLoIs6Bmblu87nU69XofahRplyVJ2YrzN4rBeLd5ruvZqVv4qFx0KI1VFHppCzDmT+l98ks7ompGRkU996lOEV6RSqd3dXQpTra+vU6d1cnJSmfx0FHtfX2qvvHbcf3x83I7nQz3s1ikElIw4pg8c9VYOAhNTGIHc9uRNLRQKgDlra2uc2WfOnBH0REgesHXsuJFrm/tEo9GNjY1IJIIULpVKGxsbvDsVxMPhcLFYJHvJ5uam6gU7EfzqpA1qsDBon5lSLmwOAPmacJxOTEzMzs6qsLeXCV4oFGZmZjhlMZCbzebw8HClUllYWCgWizdv3kRo5vP5UqmUzWbxWz777LOFQiGbzd43SmL34xVPnTpFyMLW1lalUnnvvff+3J/7c3NzcysrK4lEIhAIkO1rYWEhl8uRHxXvqCQI74KnV2FK8PlIC7G1tQVCAswt9xKvz4FECJuvaD5J82rfJ9m5J2kf133UdCskqexOlVUhZ4aiTJQv0FrnAjrQiBVfY53GgFSoWUrHxs/R1sXtUce83EQ1+09Lcn1QpUWwKZ7rbDbbbDZTqRQBJvBIsJtAx3zpFr6OBbvlGCbv9c7ZK9Bdv9I11mHKAiUYgdyeDJzNeeQ0rS0bd++wWWwFeHsOqT82cbuyY7NVKGmILE4kEm+99dadO3dIkKTqWUry5/hwOHWUNpZlxN14LoiwcuSrw5CCmRTiFwAllGfK2lPsVaf4EJ+BCKgfiMJORAZLYnx8nIAjqmhvbm4ii+Wa5ixBg6hUKrVaLZ1Oj46OollQxN0uCWQfwMjp06e//OUvf/3rX+cVhHLqHYkBsTi+N7DCq2VzZTAYJD4bgwZVCOH49NNPX7p0SQo1jZwSpH+rVCrj4+O/8Au/cPO4gSadPXs2FAqRmJRco4eHh/fu3atWq51OB/fjiy++KLU9l8s99dRT2WwW0D+dTo+Pj58+ffqpp57CT7u+vo4tsry8TPrcdru9u7sbCoUIUKICAwmnQDmArcfHx3kj0o3KmeQkXGS1OOQQRlKFHbw717uj+2gDvSgJAydrffBoC2UMGA+8t9arLladWd2WxcbBZtcVBCHCNRSbBrddxYnYSohmu9IcHMmbdMyuYRuJojyaTpVXW5JJe8SyD+6b5JhaMsBjsVin05menr537x5xXIg2VXtCPekzuH3064+YN9biOxgdylZuB+6hTQePM9kWM7JIlqUc2eq9mNLEQIvqRO46mGqdTgel8tlnnyU3kE4LwthIJ0+sIFmomCHeDghe/ArelzRvKMvKrELMnmpkAGhYS8IaJToSHLaM9SDZX5FLhAIu1DEIh8Pj4+MIbjIm8jhVbMAMhwuhaBE1x9C+evXq6uqqpLmve0PzroWqL73huXYdwqEkBoTxYcQuXbp09epVm5zTgVai0ej3v//9995772d/9mf/xt/4G3/v7/09EhKRRmN2dpbQ/MHBQbJXAwcxUxgxu7u7y8vLIyMjL730UiKRyGazJM8LBoOFQmF8fPzcuXPJZHJmZubpp58mSrBy3O7evbu2tkbx+FOnThGSQ8I//Ir4Bh66zvtjx5r0x0iU4bWkH29fe0EYzaZzw8EPRLY0Np1P9rcWU9ZvEdNOjBvkAqSw9fVRAR3r06ny0Us5sInyJaYd6ouNYnHCNawT0sZa600fYNYO0M5vpFPjl0CHp75Jr/zFVhU9yTlpvzn5TDsapZot4XNys85ilLaemc4366PzfRELhJ09exbRya6jhl44HH7hhRf29/dTqVSn0wE7Rn2bnJzMZrORSIRac2JoYOSyRDY2NiSshWIDm/K9ZoRcbirAZs0F7yJT4WRLurKJjO3LcpajsrGsO53Ozs4OaZ3RUs+dO0dyjGg0Cg2/2+1Wq1VwUtWR8pXCgUBgcnLy9u3bort5QepeK4FcSDSQXMeDz0bd2NhAUgut7na7d+7cWV9fn56ednwDnGR7e3vvvPMOQNb//X//37/yK7/yV//qX/1f/pf/BbuBatnVahXC9eHhISg89g2vD3ZcKpUwOwgIhN9NplCA4wsXLiSTSZIlkWSKoT59+vSf//N/nsrWW1tbyPFXXnllcXFxbm5OeR1O6Cf3OntsEQ/vwrYok3OTjxdr5okWhnUSKjjfDHsCcJRe2O5WW6oG+AKh7KRFBc6FHc/aUCFjtA1hUF6rQikW7JcWr7Nwh219nHCSQqhf1u9yX+Ipr7nkTjAYxHMNgolmrUpOYjJ8xOnp5XrWP/vgGKqp7KXi25oRDhBsuQQ2T4IlhxHaC2sNn4OSprfbbabW/hbhha/s2rVrk5OTMCjBNOv1eqlUCgQCTzzxRDKZGhn5aaozQF7Ak1qtxtgq7Z/q23v3ks5/Xkr5Wm24P3JceY0dndGJgBAS4pDh7BqVaTU6OkrC2Eaj8cYbb2AWnD9/ngoslATTcY4CqyonmnTn/MMFQsJoVQBRuTXHAeuocoq+c3zFFiki5R5lcDkUgRqr1eo//af/9LOf/ewv/MIvkCFHI9lut7/3ve/duHEjFApBTPy7f/fv/r2/9/d+6Zd+6Y/+6I94F6IiwX+bzWahUBgZGbl165ZKHAgSPTw8TKVS4+Pj58+fn5iYmJ6eFoI0PT2NI1rAFMzCbrf7J3/yJ++99x5Uwk984hNf/OIXySezuro6Nzd3586dW7duWQqBhfLtUWcXj60G20vyeoe6DxH244Kee4EtUEt94Y6jDzpg96NwQoL79SurzNnCJpLRsIOEXzu4a68O+5YZw9BnkG3YvYIVJYV7gU7e5z7Il4k4xlPJpdhfPEmRWjAuULHtY0RVQdAohT9/Ba2H/yARLJhGnAcebfPN4wG3c881OLhla2ez2cxxo/g3zUtHo8NkpOJ7abiC+RlfsCrJa3Rnke3QzhRZxIvX6/Xl5eVkMnn+/HkpnsViESoFHT42jdvV6n30ANsK+b6zsyNHnz1L7MZz5tKC3fyk2+0SDk6HSbqUTCZtStheC11giI16F0wmqYpKSPFs0mJUq9VUKlWpVEg4Vy6X8/k81LT9/X1S3F2/fp1h9E3ja8OjJyYm3nzzTS62ehNpFLXoBUyBazsLnXnXb4VckeWDzKVCUdLpdDgczuVy29vb77777vnz52dmZliBP/7xj99+++2VlZV4PI7Dhg38d/7O3/kf/of/4bvf/W7yuJGhkEWeSqWogkhhh3K5nEqlJo9bNps9d+4coTfr6+v37t1bWVkBPUulUmNjY3/xL/7F2dlZS9gaGRmZmpr63Oc+t7+//8orr7x23F5++eXf/M3fTCaTn/rUpyhwc+3ataWlJWogRKNR9H1bmd6e09bcdvIZqFCAtqdFV53lp3TqtsMWdrDCwVu9E4KjMjUKxlQ9cgfgUqaB/ePslVKQ7eGEPxDcyUteoDkC2l4DOQTjT/RNACgWZK8gda/Z4fgzvTRTbV4HQumVid6xIR7wBxVuq+HA8mXsUHPIlKhcdLqvgxhouJWVH5NQ82cXjXgRVLm2vBYrWPVWUKy4fzabJTH07u4ucM3U1JTMB/vaVnOxA8e7s+WYfrznFJkV/0aWFx2Dci5ZNjo6SvlqXEyk/kCnJoUxmd4U82kDBBzEqn/1Nl+8TKwAXkeL1YuROVNmE0XZuB5HmbXfYA3IhUU931qtdvv2bcLQh4aGtra2yP4zPz8PDNLH6KPl8/knn3zyxRdfVK1knR/K5NvHi6U7ixCpsExvCSU8wCMjI7u7u2NjY7hDSV1SqVTK5XKlUllZWYGtsbq6SvpQlIBKpfKHf/iHf+kv/aV//s//eTAYnJiYAOoplUpAovV6PZlMElKYz+f/1t/6W5/85Ce/9a1vvfTSS5lM5nd/93eBs+ktKUG+/OUvz8zM+IK/V65cIeCoVCrVarV333337/ydv9Nut6PR6Ozs7OXLl3O53Oc+97l4PP7GG28wNbCqGTGJY2s29QojdGwsO9TWQyNR4CiSvtmBbLCJXXgwiGzpaoVT+0JwyiHVOI7MVJEzbUCUEq+9/qgsFyXpFq/DERreaXqo9eDNjOrEBmpZOsvbahtq9xMOWAoENiDadDwez2azwrxxIoGf6h20EzQxqlVMhOsDtOWDidHm16+wYnChEIATDAblaXWsKoUeKYNoo9FYX1+X8OVUtLtUi8wOtJJ/KjmvonhJLQ8BUy5HaWQ4J3XGULsrlUpduHAhFAqNj4/X6/V33323UqkMDQ0tLCyMj4+LtuykIjt5cuf+TcLaBpE7mc8cFMgm8tZ9NIPeeHebzhh0iEQi5Akpl8tUDVfW0NXV1eXlZZu9wCloa+c0k8msrq7evn0bm0kXaLco31v/xrsI97cRobYbsCzOnz+PfsdJA2sF4JvMvUBJ6+vr8MeJQ/nRj370q7/6q4DdhMKTU0WZT7Aw0un0r//6r2cymX/yT/7JSy+9dHR0BIhBkc9isaiwCCeVqHYsD52env6N3/iNV155hQh4ZS1eX19/4403CF+cmZmp1+tPPfXU5cuXJyYmVlZWbMYC1RHVBkFiWkaTdUHbInkyf3UZWKrWmKMQ2P/jg7X+LWlyaNBKVmGVM7abhcIs7jdgOmy/VOEUJ7jvURvb/4Er7wNIwGEl2eYooL0u881j7PXt6+I+CMzImTNnDg8P5+bm8A4hgAh8SKfT1DHjx+fPnyeeDbub3ytqk9A4qGOIMMZO2KseSfE3GgsO1i2HCa45mTno8trbkByVUZd08pyurA/SUTpGn3cxsR+gzYrOzEKBqYNZJAImyS2dxceiGR0dJeEfUXCJROLMmTNvvvnmtWvXLl68iNcI5zIs9T7uWWcie68rdzVIHbDf65/eI7oPoGkT9HgD9628Ju0yBx7Fq7rd7q1bt+Lx+J07dzCVoLXZR9h3pAUCgY2NDeBdy+oRndSqyd5X0K3E+bPljJ1E4SzO4eHh3d3ddDq9vLz8mc985uDgYGlpCfusWq1SKBL5Sy0o6KrJZHJsbCwSiXzhC1+4e/duo9HAZCwWi7VabXd3t9lszs7OplKpz3zmM++///5v//Zvl8tldJH333+fqEVA8729vXq9XqvV/vE//se/+qu/+pWvfEXFQvXuQ0ND165d+1//1/8Vxh6LnLIGZNwOhUKUndvd3X3vvffOnz9fKBSevPJkIpmQjJafWRClnV9vjQsrO6w+qO9/mlTI6Afe/RWJRIRs2PkCc2ClaRdT7E1+Gif9g9XtBj3uKNHyvA6eR22KZxE31AtleA0g1TX0Zjn2wo+94E2n9SoOeT8d+/Dw8IULF2DXkkBgeHh4bGwMq1ZAFc56+MLWKSdyojxR7XabVSImg52zoaGhVCoFAgWHgbwKKLmAblbu2DJammCrpwu3Ui0oX/KG8/7VarXdbieTSRJEgFLRW3FUpcsjm7gzKXE5lrLZbCgUUpkuwhDI3JTL5ZrNJoQ80BUktdc6c4oL2z6fcPE1m00htr5eC2teOY/2Kq0CSazItp2UscVKmJqa2tnZwREdCoV2d3dxJ1jzyHmQ3W+lUml8fJwt6s2KoELDCkawWRekDzp1kOVPdwQQTIxsNnvmzJnV1dVCoXD9+nVycVCaslKprK+vI0+3t7fj8XixWBwZGYnFYsPDw3Bdzp07d/XqVfw6zWaz3W7XarX5+fmlpaVwOAyk8/LLL9Olg4MDssEgzTlCwBhB8F588cXl5eUXXnjhypUraEudTufGjRvf+ta3vv3tb/MTVWbB4QnkzTckIaG27/b29vXr18+dO3f+/PlTp05R/BchGI1GVS1PtgIXONn1NKryhklNUa4CFSexGi6RukrBwWxipmveKdyDGqT9qyPENw+MXcaDHz4/+Kw144UQH0l2q/qKMBZ4U15nknBtR/+1USCyG3yFj7erTrSnV17f1/eJasW8ElGBaEBSCgC0SaVFnNnTz8v30umtlJu2f8VikYBjKkstLS2tra0BlNu5tya5vrQZuzXNlqzeJylrL0aBNULF+LH3cbRppXgm/LdUKqFpAobW6/Xd3V22RPa4CcVzNALnxHaE5snXmWzVXhq00xwnnm++Q6dEg41xt1qYEi5iHnHMs05sPhrvFDjzwiIkQsEZEMUQsZxsbIKvC8ueKJaKS5uent7b25ucnCyVSqOjoyRO0rSq5hlpuGFliEhwcHBw48aNr3/9681m8zd/8zcnJiYIYqJ+46lTpyqVSi6XOzg4+Bf/4l8oEBn9hgqWyHe9Gty+mzdvgvvduXPn8uXLzWbz5ZdfLpVKt27dUmk6NBgBgMhua6WRhxYpfPPmze3t7ZWVlUuXLmWz2a2trWQyab2IqpMrYe0g19bNI0lqCc5WyFphjadHVFqUNu92s8eDdz2cZAEf9XZdPDauaL2m3ptYL5EVrFa4OSuz1x28bljvWziOR541srKywvOs8k9wII54EgChM8pGIArOHsXePBsOWmp1JQwiZdosl8uoAHYF+FZLs8EazmRbN9RJ5t7XP+AEHHsxVm02ShyBQpIIrVgsYjufOXNmc3MTS1CVKaym7Kv4P7Q5AVRqKP5ISTsazsv6fu5DyXLqZNsUZdq9WqyxWCwQCGxtbSnvPiKmV0CtbYSWRCIRqBcnGQpnWHy5/5IysseHhoaef/75b3/72xsbG5OTk9vb21TOhcDDMOJvRC3KZDJw/oSxkDTxm9/85v/4P/6PU1NTMzMzV69ePX36NLUmGISXXnqJcrGyEVF9APdwX2MigAlEo9HXX399ZmZmaWnpe9/7HgbK3NwcwUeKb7C6nsidemsyBWI6jI6OFovFt99+e3V19Yknnrhy5YolGzAReETpntV8VfJGkIX3+JRDz6HzIyIwozmiFFLrrDrLlXZA7ZPM+FFvhcwRPgOP0hhAGe7O3bz9tPe3iTd8N6mX4OD72VGJHFfk/TPQxh8jgBhxAX94mb0VpOyg+GZy8LrmyQAn+4KQAfR3S3L0ehWckXJOAjXn3PNtXiPA3tMb8mcfzbuAdEuTXVlZoQAKu7FSqYDaO0kGrBbsiz+cvNki0Gx7hZU+krzrpUd4Dyq5WKXqshsFNcRisVQqtbGxIXmEA+2hfajX65cvX47H48g4bzcYQwcG8XbVF34V1QxM5n//3//3fD5P3PzR0dHc3ByYeygUUloF8FM0cdwYBNFEIpHLly9fuHABmvnq6uqbb775B3/wByQDefLJJ7OZ7MVLF998800im8lzQmUlCoZJFyMcRjmz0un0+vr67u4uB78E4tDQEEkU9KaWKC20B80aPuLU1FStVsvlcoFAYHd3d25ubmho6PLly7FYzAp3adYKsGDBWzANzc7mwWAKGB+xMuwxQEk27G8MFBKVPHQNPPSao4/sh39oE9DfK19Hr+PEdxM5HKQ+z/VdzM7nB5q1DUrUbsTT7RjL9mfek4HvWWrOk2wWUCU1JcsB7Eh2giiisgKcio7e47dXtTDf6bcSmWQL7XY7k8nIPw4h14J3VlXXz+kYyRmgolPBa3l5+dVXX8UebDQa2J5EgVqUwxKlfafEjqevmiAxjT5IklLlXdGIiTTqxI7L9+BUNeRKlHTnjNFNxMeXm0jcGxxu4XC4VCqpCgzLyXmKMzubm5u1Wu1LX/rS//l//p82sl8sMcv8ddIua4LARpWsXbxsOcC5Hncf0oTag9DvwFtJAqz4HYBsUnjn83lqbn3ve9/b2trK5/MK9+92uwsLC6+++urzzz//+Rc+Pz09Dc+MpALE6Nt7Mn3y5aApE8dAfA05Rur1+tHREVxvh3al5HAcAJydaLXb29vMbOq4LS4u4hqF+wRyRZpy4AjBlYrtQsgSkEINSWI1kdRODJ0l+XB4w8Mhj40SYSprnZVr3m3rlS1eh563fRSnom1SLOSYFfLuW3CyvzXvjcO0P3c8BL6/shm02QIu39tR+3uJDF/JyBK0olYX2Cz+AvKhM3u3n+NFfWxsq08DA0UhZbHiM7RqQi99k02YzWbT6TR5fCYmJqCjTExM/Mmf/AmZerweG3sfJxlTr9arD3KrWr4q55/ju7fJdvk5hUW8oQ022AfNzmv0IQ1Vcl4kKoW5o00rydlJ3pFAib/4F//iv/t3/25lZcWqNjbuv8+trJtBSDF/soRcubwQc1ZjLZVKEEVIFkgq4EQiUSwWCYGJRqNPP/308PDwt771LdyDLAMO7NHR0a2tLZzMFF6QQiqOKcvMUncgmYhcAXyMM5PrwSG91oZFqPhJPB4/f/58p9Mh6lhn8+nTpxcWFu7evZtOpzmN+JUtkqv0Bkqlz/TJFFAcipfMY8PBfTNZO93+96AgP3azUs46PGzSVKu8WtHkTQbr3NMq4LYeo3Ol/T+Hrjg8D2ow2oscYd3/n7Y5YZd2h1gIGPgMKYkSoVh43dki170m2DE9fI++Po34EaEZNnWA9yn2vUA2ARap7trtdkl9RamUmZmZO3fuTExMCP+xfeOUkjnPl775GXq9sv3M3aTtksgJqwUIW1XGid7kKcIibfCRRbdswWybK0MV3TRZDpgD9CGOqpO3t1frdDpra2uf/exnL1y4sLm5qTrivvag77AIDpI2jT5oVTmuIX8WL8LPZaOABxICSrKIlZWV0dFRysfk8/krV668+OKLa2tricR9bhwtk8lsb29PT0+Pj4+TpxdXHg28W0mlbbycdd7ilMtmswTstVqtVCoFrYg8a+qqo5ehwxItyfUsSF3fbrcnJyevXbt29erVra0tm2hfo6f7a1miOCv9CH2wR6D9rI45KQrU+pM0+kiSPvM+2GOP+2ri/ReP7/cS05C4fN3y9g69jqhemjW8F5vz0itgLRrB7vtpHRo921sXudd7enVnK52dNKey0MXSxYpE1rDC7Pg6aat8h9iZDy/w3wtjQnGQZg3G2l/KK7ofEw/kDgOzWCwCfZTL5Uwm84lPfEKKqrohq0Il1x7VE+L4mhWby2KSooeQGhm5T8rEY0Y+KfVBEUNsTs2XgGYHtYdPCePKCnEvKqeAT0GZJwlSaDQab7311n/xX/wXn/vc565fv4757xyc/avDaAAlr20wpFWuU6mUbAJMXeYUeBpFPpFIVKtVRVtQFWxiYqJYLP6Lf/EviAbiniT7p54WdbnK5XIkEqFOvNaVfL+co3ZMKG6LSzMSiRQKBao63L59m3wpABTOOOhwYnJHR0cHBgZ2d3fhI9k5SiaT2Ls//OEPv/SlLwll7rVfMDEFhQnZ6JWZz4qYx0jd9x9hO/qw7uwkj/UiHr4YiJodEwsRf8htaBy8zn1QOw4PD4GhHljQVle1fqo+6RP7e2+diDXV/ojFYqgq0JXAGR1J7X2QJXI53zsv6dW4fQei3W4nEgmEtbQkp8P2s2psU05pa2srEAiQ15jg8tXV1aGhodnZ2XK5TOVpm1HXyhrlL+XEsswzeoiPlw2jKFCqYusm2ATI5Xw+L0DDZgKDYEOeE+ut3tnZQYg4jwYZQGAxTSjm7Xab651ZEG2cFwQRcuZCQt+W+3NaKpWan5+/ffv2s88+Cw2JoGrLNOivZVvQTIuNiuBETBAckM1mbT4pLAbe1z6L2i5kvh0aGsrlcmfPnr1y5crf/Jt/E0OQ6CfcFe12myqIrVbr4OBgd3d3cnKSCBfgBR0VMqqkwIZCIYKJ0ACUZRf52zhuxIn4jgD2wZkzZwjgIidJuVwmKZWgCVQKSqwRYU/CWLIi66Tn/6woatITfYN7hohc3zG3cKU9YgXRWMOxv9bs1UCt6Dx8mInWS4Z4zyT7J0tSZFNgNHMB2xDeC8BprxHQ/70Qh6/urM+i97BacCmRaaBer/PcB/msLUDupOPq85L9h8k7ZFZVB0Mgl4IyQPXyNviy3PqjY17unVcxRy11NIKHatbI0LGxMRhXnU5ndXX17t27586dq1ar9Xp9ZmZGafB0xgjyw12ueFZnOeJ3hZlgWTTKv2wTVBHgoCh5NWHNtkyXl2akCgYO31M0ACqLK1ushJodKC8cb2fn5HSX6nH79re//V/+l//l1NQUqqioC97TzjZfCaLlIa63V+fQ6kJXsDEIBNPiY0deP/HEE//T//Q/LS8vk/+LgFUc1JL1/P/NN9/8tV/7te3tbdx9ehCHq7VHGflz585tbGzANL979+6ZM2egW4jyr55bZUIs23Q6reMczB2ah54CmpFIJPb29j796U/jVpXsIGjLy3ruP3EfMSX9v/821ENQWD4b8whCi3mq1BocOdheFnWwyaHsI1TtoY8Es9aPVSBkFlvfxk9BVCf8oddu9AWvT24OOGNHJlJAsT6Qy6MiVvaH/e9GBhLnJH/oGh0eHkZpwl2TSqWIa3jxxRepRrixsRGNRkOhkAMmyKEnfzo1CnTzdDpN9QBlzmIBwW1XMScuVmEnlRZst9uIZjs+wjScNHUK+XUWn4XJVAhKmqBkh6XKaGQsrtdrvnoNb7vdXl5e/uM//uNnn332l3/5l3/rt37LArXedFQnNLclshWG7njqwAG96RBAlgcHB5PJ5NTU1Ozs7O///u/fvn2bFJpE/ELyaTabGIuUqYUa+OlPf5rET/aeaFtCaVqtVjQaPXv27C//8i//z//z/8yzarUakaiMtorjeEEGSdtsNqtauqptXS6XLZMBdL5cLkej0ZmZGRZwo9HY3Nxst9ulUolUCt75cpovcuJsXl82sT2cHGv1oZPohVmOTuyi7BUKr+ZQGxCUQFKW9Cl03skV4asva1c628RRa2waALv1gDdV5FYg3oNsP9aTLjeCr2h2RtD72ZlLX+xCPDk2f6/Z/YheYyeVhDfGWoGztgyYNz+W1ewUppjJZBKJBNWpg8Hg5ubm008/TfrjiYkJxBZwBHN2v3bqB0MBE4On2KRFmK60nZ0dSWGSP/DZodOJ1kNKANV/890GNhRbXn4nG7iD9dtwDCua7bludQdfVfok84hkvHPnzh/+4R/+4i/+IuSQer1+woDM/s0m2XEGUIOMCqxxgIyRSCSmp6f39/d///d/f3FxsdPp4KQVkYM1nMvlSHULoFGr1f75P//nzz//fLFYlMaEHanVVS6XqWD59ttvv/POO5Cy6/V6KpUi4wdcb6Wv8o2OQ+NJJBLQkDhZkcsqwsCOhgjYbDb/wT/4B0888QR1IOEX/uQnP5mdnfXaRjrFH29CP97mlTBHj9IHLXXfP1nIIhwOK5GTqmohuKV320crgaLzf8u06RUIo/vISOJL0tCrxrGE0gNh7fC37L0cR4T9k/JNS3fT4nCcS8gmoiKHhoZQIXGzNJtN8mr6bns5hbzan71Yp65dcN4ydBpNhGkul+O5IMh4jRhlnTeyKPkyHA4T98Vo4nYHVm6329lsdn9/PxqNQru256qNzJTGSnZDPev9999X/20hAhtPpDtoySJr7Mv2clZYaeu4Pe31DoAjLd4JLnUGnD5rpjAsICn3yi9s1xKkyf39/e985zvnzp37whe+8PWvfx0OmVaXN+dvLy3eQmSCAtgPlUoF0NnWfdfGIx3V1NTU0dFRoVA4c+bM0dHR66+/fu/ePfRo1eIYGhoql8vBYBAmMnHkJGscGRl55ZVXfuM3fmN3dzcajcLMk14i7Z7hIucJVGsiXRHZnBZq1nUPbQlvJFHszAhrEqFcr9fhpcDv1pzi9gD62NnZmZ+fv3v3biAQmJiY8KpZuA2Yfbrn8GdO2NBVHxrI2stF6euvGjTmnVVEHLnp2CLKl23RD4e+wuQ6MRD8RPXzxOXv1eeTRCpqH9lUlzZA2r61mJoP3AtOXitHydULaKyth9ppVsOSnMV4VGI8QpsymQyFom0aYq63yZscE1Vdd/KT2MGiSSXES8D6Jrubijyh3qqagYNdaFkQnQieiMMBUU6u4f39/dHRUeLfQFfYnLxXrVYjtp70cppyiy87B4/WonUMejdJr1QG/f0HzmB63aq+yksf68qZApWOdNASX7NXqhxZv1977bVPf/rTZ8+evX79ui2x5lisj9q8lpPDv4arhzScmJh47rnn5ubm3nrrLdJMohqjZMitiuvFVl5m792+ffuHP/zhpz71qR/96EfIYsLQ9WiCU7x2A6HCW1tbEI1gE9ryVMBleB339vYQpiq5LcxKy94qUqgXOzs7GxsbnB/Dw8NTU1NyMFqDXV4HhbRwHgiFt1rdSWS37uwFc2gngVWHPqyQaQH3Sv/kfK/H2ftYOWafZavQ+dbi+ejsF9/oa2dgnbEd8U0x7u2T10fXZ8c6uj0YH94PpjyRSBBXAn6HaBOu6lSZorSPPXYk3HV+OpPKLuJZSqFLkCTeIYJ3ZOsJj3ZWg16H3PO7u7uEPo6MjDz55JNnzpyp1Wrlcvns2bPFYpGyKSSCUKKJQCBAtT260Ww2rUuw1xj2ifej6QLnEO216PuvFWuseDNneheT143h0ONarRYVzRWI0b9xLO3t7b388ssHBwef/exn6/X69va2KuDpmv4o5MmbFHYt8nq9Pjo6SlUXwhStpMY+U0YUbCwhwhqHRqOBifC1r33t1q1bBNA7BVBwFEPLs10Sjol3FyXastptSSPgO8x2VYMTBC8iph0lOKbpdDqbzRLy/vnPf55c8I5yxjwqRTB3s8lnHupMsve0of8OhO380GtAOz85+rCbhIG1h42DETsLzEv8twpTryLavsfGo6w1d3B8vS/eXCLe1d4zgpHXloruaHAqcOlkPrLRPowOQmp8fHxsbCydTsPbR9u9cuXK1tbWxsYGHmqoadwHXjDPUtCHHudNtGYHAkmNC3Fzc1O6vIq/AIli0pI5z8Y0S0BrXtFQYrEY5CoSHB8dHX3729+Ox+Pj4+PvvfceBU/tLgINiEajY2Nj7XYbJpaD5jvrzL6Ft96Ng2D2UTNPooHa49ax+/pI6pPcGYqxSsr2usybARIJ8tJLLwWDwV/+5V9+5ZVXKAwvlNaL1/V/u4e6W+xWPH369KlTp0ql0ve//31kN4gBchadWiGjWFe1Wg2uBTeEfwI68fbbbz/55JOvvfYaAl0pMsTx9woUTnRUDfQJ72jv7u5Sg1ihNxLocgIrktNaJIg/ysh1u92f+ZmfIWmJkoJ6p9uWDZFr1EvL6ePpcYbdaosPFda2+UZED39Q3s+yKezZaZ3qqiti8xk4otm6sr0qoFfP/fff7s+6gqyQShZeIGUBEgrbLRQKxeNxy0uVQ0N6B+MCTS0cDkej0YmJCRYTeBxFmBCpp0+fFpMMgGJkZARs0aqZdA9fNuI4FAqVSiXSUnMA0A02hpIgizuBxPSCD/J2WkmqU4HVPDY2hm4OsRqXfSQS+aM/+qOFhYVMJkMNePHtwOiJA6YuNYAPwKKNYXOUEUeAes9wR0z7xg096qqyRwUoE4clOCy6IYhEH+XaOg+ElnJieR/nWAa6D1Sz73//+81m87/9b//bf/2v//Wbb76J7VWr1SjwZgmmvuwCXaCuquqHtUWi0SgyN5FIPPnkk3t7e7du3drd3UXHJ+zTdttaclhO3W4XUELXRKNR9svc3Nyzzz578+ZNhqLdbjOkSBm5IkDhVJgCSju4XCAQINzcroFwOCwGp2JzuIO48Gxnyjxxq1arRfoa8rgqXZTQQt/Fw2hL0CstlwgSjlyDlGLlgG6ogsvWTlLcgGMdWmTJWeoDH/7MwNosRvZX3nNakf1+m8DnKR9XmI8V/f13q0KQFAT+ITYIeoEK7TgDTV52pU1AjqNN6Bn2NMtms8rkK530QYD18X9gtZlMhgzaKvKNNbe/v09wgc2WiSpBUjfZPsTyVo7b4eHh6uqqfXPZg3LiORwVby7pPjFyaBmnTp2KRCKcDaOjo6jqFy9efOutt5rN5hNPPOGNVv+4vOf9rU4H6X68m0vZV0Sr1I3+rAyrKHmLy/U5Nqx5odZqtUKhULVafeONN+r1+q/8yq988YtffO2111555RWVBvY1RHx7xWKWOwjlgLzV1KvN5XKZTObUqVM/+MEPVldXcTILIrMbQUGnMk7JwT02NgZVToOGUlKtVufm5r785S//m3/zb2A0E+8j8Sp2sxAM/kR+D1sjyXdjW4qYmC1eJBTQvNPpTE9Pj46Ozs7OBoPBZDJptU4l27I17x0s1Jt9zD6CD5Yj7Ahr/dxyk+1nBG7/wHTbWG9S8vT9Q/GZh7b/sOozy8w35e/9WcAXrLgMR/GRZo1SCTmMBAi6uw0DQRdgEFE/icAh0yPAQqlUyuVyCDsBypTOgi8shimPoD8UIUWrRaNZWVmp1Wq2Gq8V7kCNlsBoc+I4+V2dbeAdplAotLGxkclkstmsUsKjSszOzhIeZul3vewmL2XFisteZ69vc/rprWTxSJ4f/VPnufX3+j7RwTp9hbXjL31ofwTLcgwvLCx89rOf/a/+q/8qHA6/8sorrA12qbcnzlQK/lL1KbZBLpcbGxubnJx84YUXhoeH/4//4//4kz/5k4WFhXw+32g06vU6LmJizWU1KyIRyVgoFIBBrKYidwvEvh/96Eef//znJycnV1dXWS2tVgvvIpxIbiW0jXWLT8VKol7zaNMTqllAQ8F4qBeYhjAjtd5s6WRZ2F7aNdYAurwKk3vXgEbbQWMVmfnQeoa2BoVDDw0fmyaWO4GjtX9ejl7/7POl72X9McCTK0mWiOkLBHHYK3mI/euD4SMPJIvDCb2VQrGxsVEqlQin9q0lKAtRqIIlLQkUJnbR5rJBL0gkEsKVnOL2RBl0Op1arcY6htmqsF2bJY4jhyl3JlJgui96YAFrTQ+fCTPDyEgkEhTQk0qVSCSAiXzJFVrH/8ExLwt8+4KPdtmpIqqD5j/05uKHPoaaHw6HoRizWNfW1v7Nv/k3v//7v//Zz372a1/72tbW1htvvLG5uUnqTu/8qinzn8r7UhxgZmbmi1/8YiaTuXv37u/93u+trq5SlW1wcHBtbU0KNWCoU2ZMUjsQCGxvbweDwVQqhX4tCJEFAGRULBb/5b/8l3/pL/2l3/qt35IGA8CqJCpWb5AqjQCSzugN0WYhIdMlai3Io9MrkUhsbm7mcrlYLIZCbVUxb8hbL9CZiD62nl3GTkpkwU2OsFbGWtix/NXylC1r1rui9P/hY6asSspJVjx0XX3Erfex2McnfJByB3n/+oAaBe68sbGxubnpUF4cFyLNqRtv14rD1rInIXMGJmXVSW9aTnWdD+y6g4MD6hnidhd25nAGbJds0S8ri3spLFbKsA2AZWq1Gkj6+vp6Lpd78sknK5VKIpEYHR2lYi/x39ZCt12SEPdymLz8U6sn+mJ2vkvH+6W9gwWU5a/z3kTMdElqWw5OgkzZArRhCBean5/f2Niw+aH0UIVcORPkdIB0MTZjH6viBz/4wbVr15544olPfvKTTz755Pr6OiXGqXNk8+3B2kZ/jMVi6XT6qaeeOn/+/FNPPRUMBhcXF1966aV79+4xm4lEYmlpCYYPy56EGMSniH0oj5YwfWkYOoNZ1Qr4pNDi3bt333777V/91V/9vd/7PbLys1pUEc07EZLgvhNE4Dimp1MtTwC0EFLUZDz5jUaD2HQqjfjSM6THoJKrEpiq0gCHWrqb8GUtMzFWpatZTF+Vy71OxYfylA+OR1uaHAePY1X3urP3KV4b1Osl8t7BXt8LT+/1Fl40vBe2roXh7JH7SzYWixFHWywWe5nGXovYS/zSn7wvaZeFpaQ4pmuf97fbhgrrXoSul8CyHxzT0nkpm5cVkUql1KtXr6bTaXKkLSws/MEf/MGZM2c4QtiWMg+9XnXZGSfJRPMRW38VwEkV7fwV+0lnqk47i2OIewCe0Ol0KAZP2P3U1FS9XldE5ceoj2Bavfrqq+++++4f//Efz87OXr169ZOf/GQqldrb26MPkpKIaRI6IiLv3bv3p3/6p2TWxw4jOKtWq7VaLRIz6VkjIyNbW1vwQHzjCWxxXgefVaFCejIyMvKNb3zjP/vP/rMXXnjh+9//PjEvEliOBqp81iMjIyhPkeMmJQYWk+xXb3ZMK3/Vtra2fuM3foO0VoKnrVZhbT6l1XbgFCs4fB2AWiTk7cMnbHd3LzHnbX2EwP6xvukroPs0ryTxCl+rSJ3kJo/XTrgjJDNhuP90dqhScXh4uLm5CYHBgYqclBF9cNhH6uvJcVWnHR4eUjfvkX6rhBt9jCYkr7LyI4VTqVQ8Hmc/kz54amrq4ODg1VdfzefzONzlhbNooMZK9USknfXCrG1nfJe1o3Q7Z7WDPPQaH4f74Ryo+CfsNlNiGhsaMDg4WCqVGB+oZsAXissXHfgk+7N/k+5MSdZqtVqpVO7cuYOCD0GCbDNKbtftdtFho8cNMRSLxQg8yeVyJLPlT4lEQoiWN+DLt56p6DGi+tIIzeUbONeJROLFF1/8/Oc///TTT7/77rt9DmzuaenDzgziMSLoVzRQ2ysH4MKx+dxzz2E6KOrdprS3/gZBFg5xTTNoi19797JqlekM0F7z9VHbPWIf1GdfH34QrWO/7BUwZfvvxLU67nQZIr2e60hC+40dAef6Pnc7iQBUBk0BYg+ocu12e2VlBZFkowdPEuD7qK0XuH6ShvuFBDqPtP+9pYt9f24hPPRH4M7Tp08TQDw7OxsIBHZ2dn7+53/+O9/5Tui4cdT7+iqBn8TatgurVzf6Myh6ffPQurTO9uD/SCsLrJPmyTqvtEZFzFIiTfg2e3t7o6OjoVDozTff3NzctM5GX3D8UZtjFeLpIu6GilzIF7RCpiMajVLEoNPpVKtVUgVQ/2x0dLRUKh0dHZXLZawlMptrHEglWiwWLaHTaYhUzCm7R3g6izMSiSwuLsLQf/XVV3/t136t3W4vLCxY8qhtyh9EFWYsBiefCUcOlpyXw+5dOZVK5Ytf/OLBwQHcVmm46XTaV9/0frbL2AtrOJeJUcY7yvTUIeGkuX+kNmh8SI/0K28/H+PpvuL1hC/Sf0dbceE9gFGVHgjraDSKMUi8n/WunsSd6vvXXgiR+uT7uc8jgsEgRGYwEFUheiTuhESwHqpoF1VC0QrDOUNKzGazWSqV8NdvbGx0Op1IJFIuly9cuOCVuTYGDH6hJLXi9By4X6/gi6NJSfGGxlod2UGoFbsh7zyYo82NIGFtURr5+kW6l02gjDbw5YPBIDZHMplsNBrRaHR6enpxcZFoNySLdqnj15IKxuQmEgli9/VXoedcYM1tKiwT2Z9KpUQxbjQaLJLd3d1ms/nUU09tb28TVN1ut4+OjojVHhkZKRQK6Lz6P5AIzKLJycm5uTmY8nZ9OrADKrwCCPleMdlcn8vlSLTUaDT+8A//8L//7//73/3d311YWGANO4Ke4a0dN1uX3S7j4eFhcEtivuD5qToPiwFKSS6XOzg4mJyc7HQ6ly5dogaNFls8HlduPyfpoBoWoZP8x/bEqwsLqhZ/3KkUbp1bJ5G5XpTy8AONB8XTiWvVlc59epEC+1ifvf6q9dxL6FufnNMf54PduXZ12dQXKrH2INxc425DPB9KZX3s9qg6tU5sJw/DCZ8leA68z7JKIWmBIQoQF5mJiDLOiWaziTKytbVF9Hk6nd7a2lKIl2N+qufed7HLzvevzvcWg/KFL4CMaRjmKJuwxOS8FTRp5aZlnim4yX6WexqR1Ol0yAE7PDxcKBQQjhDwV1dXt7a2EDSgE6R86zU1NjakVqtBfEbPJeOoV1ACpiNbvXykqampSCRSq9XOnz+/ubn5/vvvUyARjxx5msi3tbGxEYvFcBoD+9RqNeS+1HC0yJNUugkGg7ZcoSo2oGiTb5rgw9/6rd/62te+9vLLL8/NzVH7wwIOPMvyTX2dzBxLXIlhR7ij6CV8QAN77rnncrkcDnnlJyE+Uym6rCfDN5+iffoJN3Uf5PcjejKG/Oih//5bH0XWYj5eoe8NMbf/53rISFYOYPndF9bwpbyHpJ7n0Co+xtieXs1hFOGejkQiTiqGk+ChktcE6RK+haBBOsAoKhaLjUZjbW2tWq2iDg8NDW1sbNy9exfpMz4+furUqVwuh/ExPT09Pz9vM8qepLbvScq/+p7hFH/Sn+SpR0vi1WwiKlXAQYGlqLbVRLybU+atpW/yXCF62Ww2Go3idI1EIslkMp1OE8ZCli7uCUICSUZ9VupIku8IskDFQ3CQps6GDjKMpVIJpgcdRv5SZUYrBDh7dHR0dXX1zJkz3W6XyoTZbLZYLIZCofX19Xa7PTY2ls/nAa/1FBRV3KfdbhcvjqP9eQ9jPVrymrCdcDhcqVTQLnFpQvdcXFz85je/+eUvfzkQCLz99tvaU4Im9dnBlK1GJnkNgMnUg14ipikyMjw8/Iu/+ItTU1P4WjEarAdS6TLkcnQUXq/uaVVsr9JgW/+N6btfHqrtDhqE3Srp3nAw7336Mz16fdO/+Yaky/9pp8wxnW1IlHM3Gvqf3Y+oKQ+SL3sjGrwT86jtJM4up3lxNGnB6BGsSK/r1tdS05+QWZlMJp1OFwoFEkwnk8lsNosBTnabTqczNja2sbGxvr5eqVTK5fLg4OClS5emp6fz+fzAwMDdu3f/3//3/52ZmZmYmHjzzTe9w2VBiV5c417JN9R5GTp0TBa3vZtvqK7i+AlcQl4rbZtFHvroBbYPEu6RSCSRSOBcBR2mqJUIswcHB+l0mopWMDeYKS9HmEbiZrAU4AJiAlGE+a0opIODg2SdRXVVRh6bigA7qd1ub25uUi+RTNPlchkU9dSpU6RaBKoG5kKKtVoteCzUYyOxIpaBqvdKo/ddt5LXXCPCDB1Tqc+Dg4N33nlnc3Pzueeeu3r16jvvvKM76EUsOu8wPq0SAzqhmshsfmim7XY7l8tls9nJyUm8UCRdcVL12uc66/MkST697aN7kh+7DZ4sYsU3ddpJuu3smv5OUeufdz7on15CiyO4KZakPylPy/0cCHfu3NFXvbS//qiQNIITXu/bLJ6l1wNTi8fj+BXFknayR/WyOJS2/+zZs6R0yGQypJMPBAJsTjL8ptNpWZflcpnK1ul0mkTAKF8DAwMvv/zy8vLy+Pi4HVlJGTJSQbvmAlV4IuX8/v5+MplUYVOOSZRibmWL4yWTSSSXUvGKGmFTaGqr22MM9daL6J1E69Hw4uxKJpO5XC4ej8PyVBo5YKLh4WECphBzZB8EubIlM+yKohA76Ha3211bW9PTo9EoAKsVlLIGYIaB/OBFFGFOcUndbpc4F4Y0FAoRDAnGQjAh4wPsDuwzNDREAm6SfGnMIXWIocEZg/wVtij37ODgIPTzoaEhzp5SqZRIJCYmJpaWlljDa2tr+/v7Z8+e/dKXvvT2228TiU5lGcXy2KqGfcA0u/n39/fJd0gy7qeffprkvRgrFOR0ABZ7Q7uVtLB9pbPXR+LVMb2IrbNJ6Y9zpTcXudOGPvBaW3el7b8EiLfP+hLVx5twQgveV+n2wm6quOTQ5KwMVDEBS2936oP38uEh96xYu38si67kzRztdNqBEb2j2WthnUSz7uPY9NL4LabTa3bBnRnWRCIhTxRGN2WZ1tbWdnZ2wBaXl5dhgLGsSe8Zj8fv03g7e5lMhmD3c+fO3b17lzsQQ4GYlhcUqNSeKOohCiPlaC2ejtGAyiw+xtHREXndpNMpF7b8QhbKcCSyt2JAn2myfxK2g0AkPlPZwG1woBZ9KpUqlUqVSqVYLCL4vNlrHW6+yqs7ugk+QEAq1amxJV14KBoxqii9bTabGBwomPCsATcSicTW1ha5w/BnjoyMAMugAgcCAVhGMFk5AjudjoVKdCrgPtEL1mo1WNulUgmdnQQy9JxVwbFEwq/V1dWNjY1yuTwxMfHJT35ydXX15s2bvLJvJZD+hiN/5YQ4PDw8f/78f/Kf/Cef+MQn/uAP/kCVPfTb/viD9Wc+hr/Kq07++29HH7bm7Ujq1WwqZnuxk/DLfm9TcehPgqoksvleS1qOfekl+q2dF2+5cKd7epH7PUMinCQT0CMpyw+9oI9TwnqNrbqhDWPhbC/G7dyTCA5ktI2Gj0ajgLDUJh8YGKCcMGIO6Umdge7+/bobWLszMzPj4+OJRGJsbExnwOHh4cLCAmqmHOh0gNQ5ABSJRALvEwJRhwdKvdjKqqmIELecP5tNwqlM7x3SR9pvzqnOCZfNZklETuMFRcUVqQtmAgvJuWcv7Qxeh5gDaoVCYXBwEP8bkhEQGXgHJRf1QnQRL4+z1WoB6Q4NDW1ubh4dHY2NjalWNGIaHgjHAHLfm7LZdp6Fh7h3Nh4Hm6oYg0tQcCMYDMZisUuXLsmnf+nSpY2NjWq1SrD72bNn/8Jf+AtvvPFGqVRymELO1Hj3iP0wOjr6zDPP/Nf/9X+dyWT+r//r/xofH3fmok/TKsITiy+hXq+f8OdOP51vvEBBr316Ephl0CDXNqsJzRaU8GIX1nqw19j4Jt+0EIrEloWnBDU2nRZ3w2GgtFZONVRu7puK0umVdyXcZxqRmagPGcX3vg8d3/6Gdh/934YRsvTtIEprIwTWWzXGNjxFKr4lfIDAsJ2dHT6kUinO1a2tLbnIiZOs1WqUJV1dXUVWjo6Onj59GtiExQ04UCgUKC9QLBZLpZLFE0hrWa1W19bWsFWtowxGbSaTCYVCpCTlGxWFkiPI5qCw5cpOcnBaML3/vCCYKOVDmQhJKOVUQ1qpbi9qKfWfrKdBS8UJZ1A1UicP19DQUKlUUhYBGZiEg0OnU3YwtoryTpBRD5MIz97IyAie82QyubOz0263JycnWQxkDdPOsTh4q9WiRoT18ChgRAeVGjXhgBM3NjYwR4DXwB/C4TAYEa/f6XQmjtvOzs61a9c6nc74+Phzzz0XDodv3LixvLxMtFGvqXGmlVYoFJ577rm//tf/erPZ/Ff/6l8FAgEMlJMoyPaCTCaDZDm5oLedeeg3Ts/7+E5o3rD7I4On+VJWfP+pABPV/PTCzSI+Ot2wNUvJx2IfLRjEKTdoS0izRxy5p/4o5+1Didv3VzPVIryCXDtKqqgtrSKNRqoE6pXewUHBVM/Cdq7XVFnw6MyZM5Bq4fYPDw9jdWKicg2nGftTmavg2MGmgovGO6Isk36a0QFmJZsVeKUwE+XAzeVyS0tLbOylpSUENDdPJBLVajWXy9XrdeDUc+fOraysDA4Orq6uptNpGGx4t8QmFrsDEQ/kwrsQaYJSqVPdLgI7tb44g6+3ts8/nT+Ro2NychJhhKZgq2vqzMBVvbW1xakfDAbRUsGOU6mUmB5Wm+ACMkqTPpvRIPlyNBpFyMIwIfgFCjORe6gtSFseB8LAIUFOx+Hh4fX19c3NTZiXiUQinU6vrKzkcjm8yjY4RS5cOHyg2LCPRFHPZrNAWIhjEGFQDmDubDaLg4HveQqc7oWFhUKhUCwWR0buRzY0m03oHFNTU5ubm7dv367Vas8999xnP/vZnZ2dlZWV+fl5dAWKH+IIBZVWn2PHLRAIjI+Pf/7zn//sZz9LBhX4OagpXuhDIklh5dZK2NnZUe5pe4j2Ufb7hLl5dVj7c8kQaaC+a1I3H/nAUyIhq3wsggR9+enqAOii9p3vBqFjtuJifxhKe1PiWP1RMDAyxzJA7P+dHC992n3ej01/ZRsCi3SjtoicisnaceefKrJp35NxVMEXqwXbLiKqrLAQwhiPxxVrd+7cuXa7vb29nclkdMYopafyzjDW9XodGAE8wc4ikl1BGc1mc2trC4kZDoeLxaISzinLMCnW8ESRZl4UOsBukoOvra0pPHp4eHhubo5OipihU9DaaCLViZf2qPzIk2vZDxXWqLEAOBSRQKO0Dky7AICzLeOYJbuzs2OXo28dI4463Lxww1FUlTgUl0A8HgdqACYmTSN7j6iZw8PD0dHR8fHxlZWVhYUFAK6xsTGGpdFo1Go1OB4siUajQbaAYrHIrSAji/Ko013eCCACdjs8Fp27MzMzSEBeSqn7OHGxlrLZbDgcXl9fxxbEJkun07FYbHFx8datW+eO29WrVz/xiU90Op3t7e2tra2lpSX5MKgq0O12r1y5cnBwkEwmn3322YmJie3t7X/2z/5ZLBYrFArpdJqxQtWQdmX3o5JkOTUegZv4cBKO+UdpfSAyX9Vt5IMgL+fKYDCoEjBWFOhK3cpJJWZ5ZX3+9KhNItjWuvMiNo/R7hOw8vk8sQzeP8O6l0PM4Qk8uMWxVIUsARpLKxaLKgIQDofhxukUpVnetJNuUYMuijhfgupGo1HyOVjrXmQJVD88CThkvU62SCSiUChUP0xytBjoq9wNFLtUKu3u7iaTSV5KBWtQ0FBkSIjMAlI+ZftQxJAcpNabL1FIBxDovRL/91lGviiT86eH/hWIBlifU00UBSlEdu4CgQCHHAAFk67EQKwWx1mntrOzQ1kJTkTEBJGEfFmr1YjZY3YSiQS0DaJaGMm9vT3FvDz99NOzs7NbW1uUMiCvE8oyqhnwNyYj50Gj0Ugmk7C5uUChvKzYSCRCB1QTmRRROCHxOiwvLzO50ri5WCSZ8fHxcDhM4uxyuUzVR7oN9LG5ubm+vv7aa69dvHhxfHw8n89fvHjxM5/5jKA/7nlwcHD+/PmNjY35+fnr169/5zvf2d7exjSsVqsTExMIa1aRPMx2C6TTaSweIEHNBQlPnFDMXmvmoRCoL63Ce5lvoQOa+j98fOz9VJQPDB4N/JSDITKV4+7SbaUMWWegw57ydU4+RtPrIIIU6apXdpwNXmu416PvA21PPPFEo9HwtXRUGcBCOc6xQ9kXjZE4A7lcrtvt4hQaGhoi34JNTQeYaxkCegS7hWxqTAPefzkJ8QXp/ICVIW6vcAwwL+Bg5wUJ8+O5mJypVAr9moRWQnW4FUUUyRnUbDZrtRrsYBIGsZJGR0eTyeT8/LziJDWS+qDCTlaj0TdWrf4P6FVn1iDnatM6xBLbf5BZjjcbo8ze0FD7tnw+n06np6amJiYmzp49S9plgGCwjmq1SgY7SgihGu/v74+NjaGrdrvdUqnEoTI2NoZHdGJiIhqNViqVa9eulcvlW7duLS4uYtXt7+8TksrJGgqFarXa2bNnCasB3cI9yBZQ/A5qNaqJDgzS+OVyudu3byt5P2ctFBpyyZNTV6cg5L9IJLK9vR0KhfL5PNGhkUhkeXn53XfffeuttygLNzMzQx8s8Mp08OIkqAKNGR0dpSq0TZnNyufEwiSnbB6zZvcyXgcOsIemiu5VLsOrQp5cqVRAtdaYQq4He99BYtquScHuWrdOAcb+xuVjB8XYMZGKKRrrR2n3MetEImFLAViNjyrOysUOUcGJrSgWiwhlxCiL8vTp08g7UY9RZ+bn551z3p5shBXwaBtCZmMfNCvWZ4pnSYF2QrgweJX0Q0gcGRuIyiW4QyHCxJF3u91yucxNUPR4KNZ9PB7nmEHTQXkn4UM8Hs9kMkLzSb+gTSIz2WbDUSkcwqyV9uQkuZkeSply9GXne1+nkPwQt2/fPnv2rBjEGnbeJZVKkQwTVy03p9KK8iIBx8PcQO2dnJxEko6Pj8udVSwW19bW3nrrrW9/+9vgFcJMLKVHGCV9ELEd5UDzThwKh2I+n3/mmWdGR0e/9KUvyeSfm5u7fv363t4e7uVIJLKysvLee+/Bpld9AA4JxDSu5mazGQ6Hx8bGgsEgQNnk5CSZrxcXF69evXrr1q1qtUqdUkYjn8+T1yUQCFCCrtVqbWxsRKPRixcv3rt3L51Ob25uHhwcjI+Pt9ttMOhTp041Go3Nzc1Go/GTn/wEAwUMhPp2wDKYPhxdqD5iN1qoTQ55W60Nw0IV/0D2FFHlhMb0WoR2RXG4IlhVJExEW1t7iA9oOfF4XKeFXatWFHp9jEf874M+eFXmh3JL7FNO3hwnvNMrB8EX9ORQjS0s0wf397b7+faAKTRP9gSQtXh0dLS7u4voUf5JLQtZlLx8sVhcXFy8ePHi6OiosIhyubyxsUF5ZgfBsSeSL6zpwFhWxXPeXLKG7zH34vE46o/kCCYtmxzvEPELyGXYAk4ftKRyuRzq1dDQ0Pr6+vj4eLPZxAtHxmGOK6XD97X+vPRnB0o++YL72BVwVdLrdDrLy8tXrlwB2ZDOiEVZrVYRx0TNbG9vp9Pps2fPbm1t8dZA2Pl8HvgIP1g8HieI/4033oA74SxTAY4WaUUrRHdGs261WixXVRzHd43Apdxat9tdWFh49913Y7HYN77xDZAQxJzqBmxubiaTybGxMRTJXC5Hvj1ek/pEHNWcr9VqdWlp6dKlS6lUKplMklAbmvbKygparSinRFfB6abPFDsnUXWlUkmlUouLi5gdeCy73W4sFoM3XSgUbF1HMTh5cYaLW+E/b7fbBHNFIpF0Ok1OK991olwxoHlkLBA3XDn++y8SfbBOfp2sukBsLku5w1cBQ7SPwivOz8CH9RsH0DthJaMTOnUe2hx7whdIcbJlDHwcbQTkt9FosM6crHvSEJ0TwBGjTIakJOpDqVTCeb2/v1+pVDY3NxH3KvzjC2B5awnaIfC+th0ppygMyoLsWXl1MPChUgWDQYQL0ecYwjABvFRHPm9sbECXhggBhk7yOT5jyIMdWdzQqVBj8bJe02NXQC/EuRcDpBcqfZLFivjodrv37t07c+YMflSbj5shxfMsX5ZUbIjMSgX+wgsvhMPhpaWl3/md3yELPjFEaFXo3XJC2ioqXqNKHHMiwuXEoyds7EqlMjs7Ozg4CElcvyXycGVlpdPpZDKZQCCwtbU1Nja2tbWF45GOkUxD+gQiGwSGkx6R+vTTT5PvCe5KpVJBIymXyyRvSqVS29vbTz31VL1eh8QCsN7pdNC7SROoMUcYsXKi0ejOzk6lUoFDKXCJ/cXhAbOTHvLuWADQRpmdPviDnC4g5sViUdxEWai90qFYRU1/UuSt5Wzoe6v0PMi4OzTcHbz/yjZtoaOt2+/3PF53X/dMn0X+cQlrZ9s6FrCFpy0u/xG9i/fvvL6+TlCA1W0lXOSHdVyrAiWVDkbOQK02VFdA2LW1NdhdHKe+Jo/ur6PSXukYC16pZE9X1ii8MbgcCmlBCVI8HmXaV1dXYXoQhxIKhbwRAZppNgb35HMymaxUKvF4nPUEwYYDrH8Kroe2j2thPepCUQTjzs7O0tLSqVOnEIjyyQwODpIvRQEFZ8+eff755zudjlTFQqHwiU98Ynt7++WXX6ZoIcgAejokE/rGzRUm43Vt1et1dOehoaFyuRyNRvFnMqGsT7xqnLJArkQSTkxMLC4uUi1Bfs69vT3CjtbX16vV6szMDFm9mF86Q/pcvIjYTPS20+ksLCx0Op2nnnrq7Nmza2trQOdbW1sU5wS7h+FHXirEPQ724eFhDFNKOBYKBWAQBDevAxBfr9djsRgwOnuK1Ns4TljY5PWu1+skOUDRBhjElBRjyspWTEy8x8yXdQKRFMypBexdVE5VLTgIXp6y0qDr8HuA8h09SOUIluXc+SSpoo/8dsd/QE/PI7U+dMBe7UENRpvFTU21HmxKNsfxJT2I1WAPSZQRwhxwdqOY+AIdXmeFTgsvoKMCVM6bWySEf+ZyuYWFhUgkEovF8CKKriAAmnAG/kQdr+3tbSqcaXE3m00lJmw2m+TaR61D4gCaK5QGCtTOzg4+LmlAvnx2FroTO/fQae6l9fg2J+IOAi8qrS8MZTt8/fr1mZkZSA46uRGdTAT0ZBVTzmQys7Oz4XB4Y2Pj937v93SOgulzWwAoKR2AKgwsKT5YSDwxEokQ65/JZCKRyKVLl3K53Pj4eCqVYokyd3iAFxcX7927t7W1hdRoNBrb29twk8bHx8vlMtKc8mOJRILbJpPJra0tkGLmlJO+UqnIm53L5UALkdqEtIAFFYvFVqu1u7tbrVZVBLJSqeB7T6fT6+vr7CAcTY1GY2ZmBu1hcXGRWNZWq5XNZmFeFYvFYDBIhIGyFeLBZlgEUqu0I2DO9PQ0mVUA9GAZWRIXInh2dhYHfqVSEQtLeAVDJ+NSu95mEbCuS5vkUxo31hKZEmDE47ew5SvJ3y3/vF3P3oQ/g4+lmXpt8Y9SUMXXcu1l7qPUetNbO7fqI7Udz8FPw2xO2Lz0Ha885Usbn+ILejw0V5a39LLd3jYIylfXVqY0loVem7RQBNTiVBHgMzIygpadTqeLxaLvyFCsQH7zUCjU7XYrlYrybofD4Z2dHVILKXKnv7DWCv7oLmM1O802TFZp23olYBFwzAh3Op2NjY1Tp07ZxCDOW1BhNhAIALMuLCyQ+Fs9UUCjgrMR94I4SbhKodtMJlMoFKanp8+cOTMzMwMGTTxRu91eWlqan59fWVlZW1sDv4ZNMTs7e/ny5YsXL376058GKN/d3X3llVdee+21o6OjVCq1sLCgIBesK70RBSQrlUqtVisUChDXgIaJZnyQUHhk5MKFC3fu3EFe12q1ubk5nT0sCSRpu91OJpP8qlarKUXt0NAQiNnu7m46nR4ZGSGhIE4O0rAQ5hMKhTY2NhRMwBwBIqk4EUH5HAz4OZvN5vj4+OnTpycmJphBmC2A+0dHR7Ozs0dHR+vr66qJbFPvisvsCwD28i1ZC9uJgBMHY29vL5fLafyVHGbg/2fb4In5LSe/sv+XH6q42L8TXmKK4yG0mKa4xgB89tnOs5yb9MetLEpuFXAvZVLB2XjSnWXB8p2dncUpxE2IYwS7RLO2QtbyQMWhBvgmk582EiY5KSkikUi1WrUVIH2HV/RwX+vPGYETGnp2LuxP0LYYGUdzUffku8fNuL6+fvr0aZtK0GlIzP39/fn5+eXl5XK5TLVATZk6Y738gBggZhcuXPjEJz4B+aRarW5tba2urr700ksLCwuNRgPqG0Awr5DL5c6ePfvUU09duHBhdnY2Go1iwN28efPWrVsLCwt4LOr1+sTERCAQePrpp/P5/J/8yZ/s7u7u7Oxw3hN0HggExsbGoLdfuHChWq1Go1EkI8sGea0Yy3w+D4KMjwfaXD6fHx8fX15elmsRnAG9255J8K9+9md/dnp6ut1uE+jImSHW9sDAwObmJoFXdkkgnTlF2u028Y3kGAFP4Od4jAgdiMfjIDwA8ZQzJtcY429L2/imSnZ0QP7vxKM6tSsxUrU9OU3ZCEQOswK9MuRjb0cej5c3t7W3feyd6ZWA+pHaA2H90TsnuaYZVSpCB6Tu5TO1RIiH+twsNu381uqwAgEdTx0YItxqkalxsSomxZcSJwyB2ATIA+12m7qxUvnxasLI7lUw1C59nTePVJ3goTOizzbXBGA97jLvIW2FNTARQBY5UrwmG9+MjIxsbGy8//77t27disfj6FCWiuDkTYzH46dOnXrmmWfOnz8fiUQajcb8/Pxrr732z/7ZPyOuCuYZooRgpUgkMj4+funSpatXrz7zzDMUvW02m9Vq9caNG0iKWq22uLi4sbGRSqUKhcLR0dFXv/rVy5cvdzqdH/zgB6+99trly5c5p2u12s2bNxuNRqVSQcXGmfHee+9dvHgR7wsIA/1vNpuk6l5cXMxkMkKfGSKGtFAo/OQnPxkeHs4eN1ygQC4YFiwbCm595zvfiUajp48bPBaS5xHMKc00m82ura0pi5DUc5wEtVqNp4DpQxul5M3g4ODo6Gg+n2dVA9MNDAyMj4+DDlnnpO/2t+lBLOfKEq58k5FKA+AEwg+MOxTFSBF2///aBh9r/z70Jx/SrR56L0fHdPQyPazT6aSOG8sC350vNOP7Vr5UB6fakBXZDvxCPUm0GBl0ErIY2oQmK+sFBQCh3GKBAriz2sAHMZy1CgXdzs7OEv3Mi29vb5NVFYwbHquX9WH7T4Cc4kItz923dpyj76PxWeNU5Af9SlxvhhFrFPVQ6fztEzWeEMABdk6fPg3pTRxqhGmz2Wy32+D4P//zP7+1tYVAOXfuHDmt0LsTiUQ0Gr169eonP/lJoHyKORB6EwqF0un0F77wBaQbM0V1gmQyWSgU8vk8wU1zx+0HP/hBu90ulUpLx61er6fT6WAwODExceHChSvHrdlsXr9+/Vvf+latVms0Gru7uz/+8Y/RCvf29tLpNAkDRkZGiHVsNBrxeHxxcRGmyu7uLqdvNBpFwiKjNzc3WSEsD+h9+/v7t2/fnp6eRqeGEx0IBO7evRuPx2GAHB4eApRns9knn3ySPLe1Wm1mZoZ04eFwWOAJmjs6xM7Ozr179yhGjDWADoSRVK1Wx8bGSAly7tw55HUgENg+bg5rQMG3NkWtV8VRLJjoA97j3DejnhPkRcQv6U0AiLShLCrYh27gKxbUHiq4ejG4T/hzp538+p/mNT227IPBILw7q595R9V7f/vNh+LKPoo2Z1VdgDC5E3HL9PqVF5I+ieHvCGvfbghKU31lG7YrtgmRYxReIkBO6KqTkVYVypWqEN+gAiyVIkoEWDBx39dxzEDlSpZgdf5vtWOLM1r6l51amaXiCNNsMjlv9K3TK1jzuAenp6fRyxgWRg/Te3Nz8/XXXyfIMJvNEu2yubl54cKF4eHhM2fOXLp0KZ1O37p16+bNmy+99BLgA8NF+DXSf3FxMRaLjY6OPvnkk4VC4eLFi3t7e/8fde8BHNl13XkjA43OOQCNnAeYzBkOR+SQ4pASRZFKFG1JJkXJCra1tndlr2yX63PYtddJKnlLa5VlBVuWZdGUKYqUxCSSYp4cgJkBMIOMBhqdI3L8auZHHl2918CAknf3+x6rppqN7tfv3Xfvuef8z//8z/Dw8A9+8IPR0dGZmRkqAIW9B9Tb2tq6f//+zs7O0tJSq9Xqcrn6+/sfeeQRqmbod7y4uNjV1SVt0YVhIj0lpPGbzWabnJxENUxEAaFn0DNMWnczPgRVQArs91Tl4IkbDIZoNBoMBkk7U8MCaM6GUVRUNDg4ODU1xerFz3U4HFar1ePxWCwWevQcOnTIaDRGo1GXywWFKRaLxeNxKC6Tk5OZTGZycjKdTtfW1vr9fiQh1SShaJlpoEJp+6vp9il+tGaJbZHuU6Nq2lbAexGjv31O9DaPjf89fWJ/waOgu0lyix0aVsJmAljyec1o/wwl8LqHvpRD4zKz5OiXCNoAcVuVENR8cTNDXPD+NV/RfExlgwiWDayhORtgIsGmx+OJRqPk/SGWCdtEMmkiwQUsANCBxQEG8Xg8c3NzkKUkVajRutUEBOwfQIei46opt5VMvcqZk7+KBp68oyrJaeglKilKfWTqM9UEARImk4yNx+OoHgqTjE8uLi6mUql4PN7f32+3200m0+joqM1ma29vp5R8fHz8S1/60oULF9g/SGZglaDD4+kbjcaOjo5bb721ra0tlUodP3787/7u7yoqKpLJJHUAjCekt66uru7u7oaGBiQ1Ll68WFVVdejQocHBwYcfftjtdjc2NtbU1Fy+dgDdhsPh8vJyzsavU94tyXAq5rPZLFIk+XwezWuwIwSnsLCRSETGGRcbnok0KUZ/MZlMlpaWBoPBVCpFvJVKpTweD3RPk8mUTqfpYgFegYzB3Nzc1NSU6JBQd0NxL6UAcAp37NhBqCHlEejZSg0a3CQ1aSzREtMMbg/PN5VKSaGy2lZUFdzY2ixqIFCr1Yrer9pPTn/IOTVBpAbe3OLYOt+2mW++tXz/Nvkn2yRucVRVVQlFUl3j4oQx2gVTdG/w7rb+sa2vQw+GyEaNN02NCUCV+l1pNKf57taQiOZ9/falxheyjQvwqlpzCQMRp5+dnSVGwztWVZbECyZ2w78WJTaSirBNTCYTHiJWWwgDm1Uqog6hCjYKmUQdKPmreMqaqKIgH3azR65nzmgOvcI9iOrKykp/f39raytf55apfpqfn798+TLstJWVFbvdfuuttxoMhtHR0e985zvpdBp+m/TNQkBcegtUV1e3t7ffcMMN3d3d2Wz2hRde+NKXvgQbD7ku6Blra2sdHR27du3q7u7e2Ni4cuXKsWPHvvvd71oslltuueVDH/pQb2/vD3/4w0OHDv3Wb/1WJpM5ceLE1772NfjvxCt0n2FC4ilj2rhNofQJ4QxshGEvLS2Fte10OhG5lYP9aW5uDtYKJH2Kg+x2OwRzLLVUSM3OzopeeTQaRWwEswu2QzTGCJeUlNhsNqnOJ4UwOTk5ODgonWiam5sbGxv37NkD8y+fzyeTSRFS1zxf7IWUHbS1tclMA7tT5Z+26caplprKYdAtsdQajOX/+rHx87r22/+iZp3i4og6zRYiEJu9fgvGWrXxGv9L9Wfn5+fJRBHTCeVAvsjFiZ6hOK1qmC9bsYassplPrac9CsEDRIKRcrlcrAESgwS5FBfQRsDr9SYSCZxl9U5FvI0uq+Xl5XV1dar+IVXCcPVQ6gGjBOQVVx3fDbKBCg1hTYgA1M7iGA7BDdU8z9bQ9majtDXbRzwp9cmCz2I3w+Gwx+ORPCr+I5D07t27SQMkk8n+/v7jx487nU7yASJvjUki6KazgdfrfeCBB+bn548fP/7tb3+bYRfnvaSkZM+ePc3Nzbt27aIZ7vPPP//II48Qr1gslhtvvPG+++7b2NjIZrO33XZbJpN55plnvvjFL1osFmr/CHSamppuv/32aDT69a9/fWlpqb293efzXblyBS3yaDQaj8dlHAAZaB3AXktxJkkLNn6qVemQWVFRgaoXz525VF1dzTYmArC1tbUrKytkBWmp43Q6cd6XlpaSyaTZbOZ9qaWUrnLZa0dlZWVVVZXT6aytrd25c2d5eXk0GiVDMDg4ODIy4nK5vF4vnYwaGxsJ1OCN8K8aGrLKFhYWLl68CN8cFCifz7OLCNJFx+SCUazwPaSDD5kD4ESmExsPZ5PCSMJKjYn/jzLlxVv619v/lc2ubbMslB4NZ8LAjVFZ6mCADBEb6mYjwDtEzP+RnjWxlc1mczqdSBOk02msoQZKV7dZwjFBkDmbppxdrTnWs/pUyBVjp9EnE8PHAK2srCADz4yRRE1ZWRnUKMadgi69vSP5PjU1ZbFYsP6kucHBDQZDIpGQHoBEphBvsdfUwnEZ6pVLIbUm+a6CffrxVwOan/tRFnSo5TUZMybWxMREMBgkZYT9hVmxY8eOWCwWCoVGR0epombZ8zFoyxCrQXusVustt9xy0003jY+P//mf/3k8HscttVgsS0tLjY2N+/bto8Hx+Pj4xYsXf/jDH7Kb8hA9Hk9LS8uHPvShpqamoqKiiYmJF1544aWXXiotLW1ubv7kJz9pMpl+/OMfJxKJHTt23HLLLUVFRd/61rfy+TydZM+ePXvixAmXyxWPxy9evOjz+dQ+eDjOvBaNdXIAOM6QGdilqqqq6CwB0Z62GNDYyYIAalut1pmZGSpx7rvvvt27dw8MDAwODlIOI7orDz30UDKZPHHixIULF6qqqkjOI2mALgpk8+HhYUYS+ofJZAoEAuXl5e5rR3NzM7BVPp8nP2w2m0mQzs7OQscmUYmjt7y8THbdaDTGYjHpRgbWB/VlswkmWXECRGo1pYWTeNZSUyOtRCXz//8dX3uzQ/o0yTvqZbNqMDgaiFKYNoI4qW6r1CupPRNU+qNaDEW5xlsz1gUxa/VACpUcCK2qwuEwLomaHpXchWqe1GpGTStY9SY1Yld6/1ENN8TzxfhK7Ql8ptXV1fr6enIvyWSS5Qc2qsJ8mvMTOsAwg90lXRytVms4HBbGHgeUCfl1PGX+JLe/RdPMgsGEfqJrML5t9rXTHAVJryrTJpFIRCIRj8cDhZndbn5+/ty5c0NDQ1SI0DjibW97G8obCIbQSBcQ9p3vfOeBAwdOnTr1J3/yJ7hgKCLt3r0bh3FiYuL06dPf/va3VUkQgDWqSH73d3/3jjvuGBwc/Jd/+ZdnnnkGc3DHHXfcc889mUzm8ccfv3z5ss/nu++++2699dbPf/7zaD2+7W1vm5ycfPjhh3EmKEqEGaKOtjgNVKjL7YONCKuBFLq05WUy8HxlClHwArQCqaaoqOjRRx996qmnjhw5Qkzw/PPP19bWNjQ0nD179h//8R/r6+uPHj36yU9+cnBw8KWXXurr60PJBM0T7lRWARXz5A9wNbDRIh4L5p7P51EnxqxjSVOpFKid0Wg0mUwWiwXGApZatBOkIreg8CkUGuoMUAJQ2/7iG7FPk6gQgEW0twou4c1m9c9x/OKbgVqryaFncahNvDS/q6bQ9LkrdgI1xaVSJDRyWpt209FzafVXIJ+UU0NywgNyOp1LS0udnZ2XLl2SUmM1gGJDloCXqaNSgkT0WU9a0LiZKkhP7YZ0NlELZ0CLmKlerxfIEqoAixAsouCAyC8KYUPOANYRj8dLS0tJH+G5IO8AyxXFcE0WeAvyk37M/w9PU7k2EWODWTE9PV1fX4+FTSaTAwMDzz77bG9vbyAQoIkPQzQ3N9fY2BiPx51OZ0NDw8LCQj6f379/fyAQeOGFF/7rf/2vS0tLXq+3pqbmwIEDSJ/39vb+5V/+pfgyoioJ18JgMDQ1NR09evSmm246f/78Aw88AKe7tLT0gQceOHz48NDQ0B/+4R+Gw+FgMPgrv/Ir73//+ycmJj7+8Y/n8/lDhw7t2rXr4sWLr776alNTE64iiQpRtRaZLamHkqGQlLIY6+XlZYvFQoeXxcVFu90u7CCx5h6PhzIrSOhtbW3RaJRKdJfL9corr5w8ebK5ufmWW26pqqqKRqN79uy9+eabz58//53vfCcaje7YseOd73zngw8+GI1GX3/99ePHj5Ohgm3JLBVNfVHQJk/LAlTb7+LvLywsMC1pclZXV4ergd8djUZra2v5VmVlpdPpBGBhbRb00vL5PFxDUqA0iqIiVK1gFJKYnEQaqfx8dR4bmzgib9Wsb+d3VVaG+i21RL5g8p/oWaOFJ2sKXZelpSXqM/Tul9qXma+/AQVoutDLF6RUD+xJ9elUPiZVUiLMiNo69WZQ4gcHB4l9qKTCDSFhIpG+7MNktKThi+AAYqwlMFHRErlsk8lE2EhKkFsTBVe+y+Qzm82ZTMZms/X392MOXC4XsjjgqjSCYSVQpA4nGl+DLlMk8fF9MpkMSpiZTAbnixYhsrT01TEa2EeTlLhu/n3rvLbeFdribCpIpX6Saot4PM6NJxIJ3Ml4PD44OHjixIm5uTlEwGkiLLIqNpvtbW97W3V1NcUsxcXFr7/++r/+67/G4/Hm5uaDBw+63e5QKPTCCy9kMhl0VJBDgvRmNpvpFFNRUdHS0vKxj31s165dX/3qVz/5yU8yuffv3//BD36wtrb2u9/97kMPPcSsu+22237913+9pqbmi1/84rPPPmuz2X7pl36pvr5+fHz82LFjN99889TUVC6XS6fTmKpwOMy+nkqleIi0jMHzAH7FR45Go9wXYns8WSjVgoTQnYAmRLgIdPPKXTuWlpZoJLaystLS0gLd88UXX5ydnb3//vtvvPHG5557zmg0fu5zn0skEj/4wQ++8IUvVFdXBwKBm2+++YMf/ODS0tIzzzxz+vRprCQRjDximvW43W7gKSbh4cOHd+/eTU0pz0U2QumfSbU69G1p/gnOEwwGq6qqXn31VRRaaFwpYtmlpaVutxv9cVYHV8ISln54KpgpEtucTW18pUH2tm+Ui7dcI6ozJN8qqJog4nTyGWkPvdmZuUeWg7xPlMwWpeaf9C4mXog+38ah9mnh/atwgVB2xFKIHypdtUjXaBQkJAFlMplsNht8OCk34DEYDAYmEEQLjKCoa9rtdunwLSJ5Uhyl9uiCuoS9wHnBbVEvmANhbtJE0g1PmLmwdC0WC5fhcrlYlnSlIWPGOC4sLMBqkj5ebEvr6+vsSfyc2+1m2aTT6erqahI+pI9WVlZoyVH0/9uDUJdSusXFxba2tuLi4vPnz9vt9snJyfPnz0PvxUdYX1/PZDKVlZWYp5KSklOnTr3vfe/r6Og4d+4cIpy33HJLcXHx0NDQiRMnxsbGSKmRdAVUBbwWuGnnzp0PPvjgzp07X3nllQceeGB6etpgMHR0dHz0ox8tLi7+93//92effZav7Nu378Mf/vCBAwe+//3vf/rTn15aWrr77rv379/v8/kef/zxSCTy/ve///jx47Aybrjhhunp6cHBQakGALJIJpM1NTXCTCfak87IRqORZgJQPkj9YX/1B7aMRjCkIp1OZzwel8a++OmZTGZ6evpLX/rSP/7jP77//e+/8cYbL1++PDAwcPTo0fvvv7+/v//ll1/+8pe/XFZWVl9f39HR8cd//Merq6vDw8MvvfRSJBKhsaT92kFq0W63Ly4usqfOz89Ho1H6Fbjdbh4orGcK0xFRKSm+uswJc1mJtbW1dEUg7KBNh7Tihlo+NzeHAgxdTPXdWESDW98SXtw+0fN6S3OyROfcqBbpDdOmNCfCC9YY600R82ufKin9GQEsDZlCiobwJvVtbFX6mUpP3H7OX//m1fmnigNofkwqNZigBfEai8Ui3akpZiP1THDELi2pIdArFgOwLx4B2IJ0fmOvlgZRkjllgPgVKoJUAEiFwKSCEUYKUaEUUpvNZrrusg0AYVPPAj2c8kUcJQgekNWk/IF/CX5xLUOhEHI50v6O7BA75P+/TLb6fJHBo7gOBaXW1tbZ2dmBgYGRkZHJyUmPxzM9PQ1dmgGJxWJut7uuru4973nPpUuX/umf/mnfvn033njj7OzsxYsXT548OTQ0BHOgoaFhZGQEDxQPRSTuenp63ve+9zF0f/Znf3b27FmbzbZ3794HH3xwY2Pja1/72qVLlyBvmEymj370o+9617tmZmZ+7dd+7dKlS2VlZQ8++OBtt922tLT0hS984ZZbbrHZbI888ojZbG5sbDSbza+++ipEDpoAgNsKpxAAF9skjqrBYMB5XFxchJpWXV29maVm42ENJxKJrq4uu90+Pj4Ofk1Om1Sz2+12OBzLy8vZbPZv//ZvXS7X+973vsOHD4+MjJw6dWpxcfHBBx9cWVl59NFHp64dzz///MbGxu7du++++26v1zs7O9vX13flyhWqFnO53PT0NI5IeXk50LzNZstkMpIEYrFIshTYUMp6Z2ZmRkZGBgYGysrK/H4/wpmUbgKelJWVJZNJKIksWKyV0GEFugTrUKFqjcVUoc6CVmwzO15aCAZQI36xV/rIVSIDFYf8mfOUvBEHSBNefVsrqf7XoMwFV5D6zi9ylLndbn3Xanmtkqj09Hj+FfIscWhFRQWioBAtSYNKy2rJM6yurs7NzV25ckVCKrBRUVinmg66jzTkxSPAHdP0WJOHAYcUUQLSL6qx5oCLBtxGjy5qeUmnsBXjMlRUVBDAMqsIG3mQ+NEOhwO/2+v1jo+PFxcXUzFRU1OTTqcxBFs/rYIaC//fOebm5tra2sLh8PT0NAEWAp6pVArBZVrB4nyxoXZ2dt57773ZbPbv//7vnU5nZ2fn+vp6PB73eDw7d+6k6zHEA8jXUoIEw6Guru6ee+6x2WynT582GAw/+MEP1tbWAoHARz/60YqKiq9+9av9/f2SrL755pt/4zd+o7Ky8n/+z//53HPPISz3O7/zO42NjT/5yU9eeeWV+++//+GHH+7v729oaDhw4EBxcXFfX184HPZ6vWAXtbW1BECVlZUtLS0ejycYDAYCAXqSraysRKPRUCg0NDTE1K2qqiIOKy8v9/v9dCvXHEB5CwsL9fX1IyMj/f397e3td955J6Ywl8ux05tMppmZmWQy+YlPfMJkMn3jG9+IxWJ/8zd/4/f7H3jgge7ubtSsVlZW7rzzTpfL9dhjj/X29s7Pz586derll1+uqKhobW3duXPnhz70IZPJRKaE4nsysawF6N6SBMMdQTeYyCmTyUSjURjxKMGS+0FHgSJ+IKlUKpXL5cQHZ26z1iSLzmTmNfiAvl2hUFGF9/2W8uHFSib8ZxrpKv+qmLK69sVT1kCOaukcnhl7nqAOGk40h7oPyaNXf2uzK/85juJ//ud/phtsQdF3ZoB+oyDsleC3oBFHHwdPRNMqTO2puhljWt2BRZtNGtcKnUiDSfEtGNNwZQDTEeJpbm6ORqPMSHqF4GJHo1Fm2/LyciwWg75qMBg6OzthO2QymeHhYYfD4fF4vF6vwWBwOBw0OaWrdFVV1fT0NNVxAH/Uwg0MDGhKdQo+sOumxbc4tt806LrrQQYQ6GN9fb2mpqayspJOV9XV1blcThIJ8/PzQ0NDWHCun6JBp9PZ2Nj48MMPI2w9NjZmt9s7OztbWlq8Xi8u5KVrx+LiIiXd586dw+iXlJTcddddt91227e//e2amprp6enXXnuttbX1gx/8YDAY/NrXvjY0NES5Bwobn/70p2+//fZ/+7d/++pXv7q0tOTz+Xbt2vWxj31sbW3tr/7qr7q7u9fX1x999FGz2ey7dphMplOnTuEms+vcdNNN0WjUZDL5fL7u7m54KSB+lGvH4/FTp06Fw2EiOavVajQaz507B58afj2+CHMSlw1JE+C+oqKiWCyGcPbhw4fZJB599NHi4uLa2tqRkZFwOByJRPbs2fPAAw/09/c/+uijgHVWq/UjH/mIyWR69tln8T86Ojr279//ve997+zZs/DqyAd2d3dTOLOxsVFfX0+PHu7RZrMVFxdPTk7ixxD44l5wqZC4IbzSLhXiCvfC7ovGSIEyjWtLDKRUeqhiGTHH0ncGh4+A22QyMcKiQ4szJOqsEkwLEltyjYwg3q5a7KMaYs26EAKiZs7TvIarFVnQLXypgq+3XmXbXHHX/S31DFe7w+Xz+XA4jD+v6Qgjt605Bek7bK7sYJprgoXKC82mqte819ytJsclF01xvbRMVVllBQdrs4HggdGqY35+nmz45OQkdhmAUk+G4UCvTm3kmkqlEKmQA+k13if7pL8Xzf/+Ivyk/5BDqk95XVZW1tzcPH3t8Hq9mDapglleXk6lUsDHbEITExM2m+3WW2+9fPnyyMhIR0fHxYsXx8fHKczr6OggN8uBaejr6+vt7SUKqaurW1pauv/++5977rn//t//+0033dTb25tKpX7t137twIED3/jGN15//XViKcpqbrrppj/4gz84ffr0+9///mg0GggEVlZW2trafuu3fuv06dP/+q//+p73vOexxx4bHR01Go3QUdxu98svv1xaWur3+2tqaqS3t9/vb2tra2hoqKurI3Jir0JXOpvN8rFMJsPNIlZOMTeTYf/+/WVlZRcvXoxEIktLS0ajsaysDEkv+nOaTCa86aeeeqq9vf2OO+74z//5P3/rW9+an58/cuTI888/H4vFJicnf/M3f/N973vfF7/4xR/96EdPPfVUJBL527/927q6uvvvv39+fv5HP/rRwsLC+fPnb7jhhttvv/30tSMQCNTX13u9XpS+g8Hg2bNn8f2bmppaWloWFxcdDgeAtTTiocwd8HrHjh0wfDGaPF82HqRume0F5wyxJnl7IQuAU4vlVZtxS3W7YBGE2vK/fJgz8K9I9JRfC3Y5ibQl2UJhgyQwr4mwRVuRSlFVCmKzG/wFKxj+Y8HPq5DTzMyMePsF+QP6gkgSL2Kt9FUVGFasPEjuZrdxXYqlKrMriL7eRusvW/9bP73ta2uSTKDZbKYuABcSp0YldGu+LpXB8XgctfhUKqVRqoJmgB1XCfMq+LVZHcpbzXRv87juZqCiXuXl5YFAYGRkhJaS0jyFGIXcGtX2uVyOHNcf/uEfxmKxV155ZWVl5dSpU01NTXR8NxqNbW1tkUgkEAigzIcURiqVYiRF3NnpdH7lK18xGAwf+MAHvv/973d3d//6r//666+//olPfAKfgDKTQCDwqU99yuv1/qf/9J/OnTtHhjwej7/97W//7d/+7c9//vPZbPahhx7667/+a7IRwWAwk8m0tra+8MILbrcbKTuj0YiSn9FoXFxc7Onp8fv9RFdYHzLk1LLS7b6iooJei9huCBI0Tc5kMj09PXV1dWfOnAmHw9JoHLKQw+GIxWKVlZWLi4ulpaVDQ0P5fP6OO+741Kc+9YUvfOHy5cuf/exnv/jFLw4PD1dXVz/yyCOPP/74Jz7xiS9/+cv/8A//cOHChampqf/n//l/brvttg9/+MOXLl364Q9/mM/na2pqjhw58ra3vW1oaIiZtr6+zp5EYrO5ubmhoQEpJfpDDgwMLC4u5nI5qpbEeNGHjAYawqkFwcAFZpvZLDMmAjXQ9djmhZWg6VJPAkl4FOIfqJ8nlYWpFabT6jU8RzYDTfFBQekFnqMqACsfA8USGWfVh1UTg//h1vYXPEp37Nghvdc0tAqNhqEcagcQ4eJoHH6mu3S817dtL+g+az4gLh5xJaiozAl9ykJeU0bMuLOYEcO02+1AKNR6RSIRSIRoy0kTAFwM8jOSC8WRRFUVkhnJN4vFAsNUmE+gSQ6HA7ddLbJQOXkFtYCv+872/6oZz+t+C58FmN5sNo+OjvJhljSuNABRKBSampqikqiioqKpqenOO+88fvx4b2/vmTNnEFmGTczIJxIJMnu1tbUkZimeCgaDABQ333zz4ODgE088UVVV1dXV1dfX9+lPf7qsrOxv/uZvLly4IG1ky8vLb7jhhoceeujJJ5/8q7/6q+npaaLp0tLSPXv2fPjDH/793/99IJfPf/7zcIopv25qajp27JhoZvl8voqKCpvNdujQoerq6q6uLqPR6HQ6eaBSkrq0tHTlypVTp06VlZUNDQ2BioBQUwsOFLOyssKV3HTTTe3t7cx8eJzV1dV0hFlfX7/tttvGxsY2NjacTmc4HH788ceDweADDzzwve9979ixY3/4h384MTERjUZx53t7e59++ul3vOMdDz74IAnDUCh06tSpxsbGX/7lX7506VJfX9/AwMDS0tL+/fsPHjyIdNT09LTJZKKU/+DBg/AOh4eHYVjiUpDPDwQCBBAwsiilEXwAOwAJFSFDjXKkLD0pAxZBV5aq2iGMBSgWWawhzrgq4COij2rVLsH08rU6I16IxyYWX8riVQQZOgOWR3x8oeipvJGCa0oDsb7VFffzHZv91hvEtv3792/2aQ594lGDUMvHJKfMk5Dyf4mD1JNoWn1rytbV36U/lmR7hGStxkFC+ONHcQpgjEArpF7TarVSkxYIBFwu18LCQi6XgzBLUS9xgNTy05uROZTP51EX8fv9cKvx9XjkTqeTrqyU5OGYgAZuITlWkLeuf+etfp7XqhaB5oFq0sgiwwYmUFlZSbCFdaPzN5koiDoUf2J37rjjDrPZ/OKLL05MTMzMzFCLn8vlUqlUPp+32WyBQMBsNtNPx+l0siXQxqWqqqqhoaG9vf3ZZ58dHBz0+Xytra12u/1tb3vbM88809/fj6Q16Yr9+/f/9m//tsPh+PM///Ph4WHodEajsaSk5J577nnPe97z+7//+wcPHqyqqvrud7+LlTGbzfv37zebzRcvXiwpKfH7/dFolD3S4/Hs3bu3rKyspqYGsyuEMCKG+fn5gYGB7373uw6HY2Jigh2dpymxF6+ZnCaTqa2tbceOHXV1dU6nE6r19PQ0+YwdO3a89tpr995779DQECoca2trP/nJTwKBwEMPPfTKK6888sgjH//4x6uqqi5evCjEsnPnzp08eRJay9TUVD6fhy551113HT58eGlpCaWRCxcueL3euro6vOyTJ0+SO7ly5crZs2fpVe3xeHAsHA6H2WzGZQHugKCCMRWgGaIIbincFWFAERtxSCEFSAh/4jOCaWhAVNkMgE8xx2pVM1clfo+40qvKa03PMNVPlx8SBUGBWQRKFUBczLEmT7jZGvw/aaz175eSH9/6y3oHdvs/UPAz4lRqxO04NKK3zBLSESrSpI6sfIVkgtTLUs4A/YA+HcwMlGukxTW0BLTh4eHynMgRbWxs5HI5JOdra2v5JLaAONdgMITDYcRWqBAjh57NZhGDlnv5xXfprSmZmjmnbo36XxfnQhqEt7W1pdNp2FokgmhKK50WkskkNXvFxcVNTU233HLLK9eObDZLB0JqhTweD+ASIpl79uxpbW3FtLW1tVEbsr6+7vF4KM1vaWlpb2/ft29fR0fH2NjYV77ylbm5OQqm4ZZ95CMfed/73veNb3zjO9/5Dpl6IZLefvvtR48e/ZM/+ZOHHnro7Nmzzz//PMnk6urq3bt3l5aW/uQnP6HvOPnwQCBAawLay87OztKWUFienHZ+fr63txeQhzBC4jkxIhxzc3MVFRXNzc0HDhyw2+3YREKrqakpk8lUUlKSTCaJKiwWy/z8fHt7++Li4vj4+Isvvri4uPg7v/M7P/zhD48fP15fX3/77bcPDAwwdYnVfvzjH1+4cOG//Jf/0tnZiTR2X19fZWXlu971LgSnTCZTJpMZGBjo6upqa2sLBoPI/oXDYVinhw4dIrKUojgJT9XJIA0lUIOhnl49xEZjuykKE8Iy9lcaScshplMDdxC7LC4u4pWzDPED1Hy7HtLUHKq11S8Kif7lHZx0jWXf5nr8v2ys9+/fL/esPzR1kFtbii3eLwiFa3zAzUZNtm6akqjkE7XyR2YDPqDFYoGu63A4CLiw+OSIs9lsKBRioldUVDBXksmkZGD4CVbdxsYGjasp0WQCUYkAhgtngAUgGpswonDkNbjNf1RSseBsE3dGpVqqAKIQMfkkvnNVVRXl0UtLS9LmQ5wj+UWr1WqxWAKBwDve8Y5gMPjYY48NDw9DS2fny2azbKjE0SRsQ6FQbW0trV54gkRLgN1UxHR1dVFG6HK59u3b193d3djY2N3dffDgwZtvvtnpdH7961+npRbhOe7we9/73p07d37ve9974IEHHn/8cVRDCZzdbjeOKjwWWkQaDAaPx1NXV0eBuNlsrqmp0Ujs46zNzs4ODw+PjIxQ9wG2yzYsBw7d8vKy1+vt7Ozcu3evxWIRMhmT4ezZs4FAAPLcyMjIAw888Nprr4VCoaNHj8bj8Vwud+nSpVgs9sd//Mff+973Ll68aDKZPvvZz05OTiYSCZFJcLvdFy9eNBgMDzzwQFtbG4jz0NCQzWZra2s7d+5cLBbr7e09fvz44OBgR0dHQ0ND4tpRXl5usVja2tqgYeBdkl+RrB08PDYqDDQF0Dx9QTPERVXnLU9ZWlmq00zsBmGEOMs8Pmmvo2+gKitajQ6Lr42q6iar/6o/px6cVi5bjLXGd9EYn+s6r5sdvzhH4DrGet++fVtc33b2nF/EWG92BvUDZP/FWGsstfjUBGUw6mpqaoi+0bIhVhXST0VFBXl/MkVMNcI9g8FAXEZ2VDxrEG2gWyJcg8Eg0u/CPobnp3rWapnl1vNgm0fBUVXnmcxgFqHqB/GOLD9JBuC9cuVIAjkcDmx9RUUFvg8HurK7du166qmnfvKTn8zNzTEU3DIwsfTPrampaWlpaWhoOHjwIKQ9LsNsNrvdbrJ8JSUlPp+PPQ9LYbFYDAYDhA0UwSorK/v6+uigtra2xlM2mUx33nmnz+cbGhpqb29/4oknhoaGQJxXV1d9Ph8l5qJUXl1d7Xa7g8EgRCaLxYJ4P0lF7lGaReRyOcRMQqFQOp1GqwjXj8BL9ay52uZrB4IEiKMCE4EIU3PY09Pz7//+77/3e7935cqVixcv3nXXXadOnUKcpK+v7zOf+Uw8Ho9eO97//vejEUZhMNbZ5XIhn33kyJH5+fl4PJ5KpcbGxg4dOgRFGqD89ddfT6VSR44cAYVvaGjI5XLSOg6XAhvKVGEQxGUW1EKwBRXhpOJc5gPxrqa5h0aZiJaSUvkiKLOKabA96END1f6UvunCywdUJ0+/uNRwU4ViBMXSfPIXsXX/UcfWtvQNgtr2g4Jf8HJFNHkzaEUjbyQUOrSQ1IojyWFWVFTY7faKigqj0ShyyfSQRqZVEvEU70gSCTDOarVCw0KQJBgM4kPhC4ioKWaaymZKeEinkOu32WxYQ1qRIrcv1f1qmancGlWR0kFcdWPVYVE1euRNNh72G0wzRQqCCPEtVoJKnwdVx8mCzzQ/P+/3+5PJJJl0Fl4+n8d0wvrAD6JFocfjeeSRR2ZmZtg4cS3ZyUgV1NbWWiwWuam1tbWZmZmKioqenh72wrq6OsSgAYLF5AmXFitDpAx28fa3vx3NFiwFWBMdDh0Ox8jISFFREZC0UJvPnz8vOjO1tbW05U0kEhUVFTCReY4UwWNk8d2wxfi8hFxcCaRdbJNEcuxVlZWVO3fuxIJXVVXRwAwUiC0K9PzJJ5+87bbbXnnllaNHj/7whz984okn7r777scff5yK89XV1V/+5V/+67/+6yeffDIWi91000379u179dVX/X5/SUkJiEpZWdnJa8fRo0dbWlqefvrp3t7eUCh047Xj5Zdf5tqSyeTf/d3fffCDH3z7299+5swZp9OJADdViEKB0FflafThVAE1WfjMGRrZCIKh6vlobIVaPaj+SbgJ0lF6s76OfHH1TXkfjSHaIs3GnzQEB71Z15eS/0cdb9VaXsezvuGGG/T1RdfNS77VH9Z/a5ubgfBApBJfPsByraioIMSmFyL+nWicxmIxFNmZT+CJXq8XNwRoWyoFeFTxeJyKdrIrSKcSOVJIWVxcTOILPkl1dXUkElleXgY6TKfTLGnqmPGwNC3HJB2qdg5jZ9LEnsSkYAXsB+J6oIplsVhAfvBtVedaLIuGdqJqrRQXF1ODh9soNQIUWGezWXgCRPo7d+50uVwvvfTSyMgIjVSkPYLT6SQQCQaDpaWl4XB4ZGQkEolMT0/jmfKw/H4/Di/sAkIlgh5psiUHbjhoKXYc+eaysjKv1+twOCorK+vq6qxWK3rNgUAARiaNz7lZ8pOS8g0EAj6fDwVRq9W6sLDgcDhErlNEYFZXV2Ox2Orq6tTUVDqdBuUEIyIeZ1pynbQG7u7u9nq94tqvr68fP36cQs3jx49DQOLRTE5OHjx4kGY39MkNhUJUHlosls9+9rMTExNMy7KyskOHDpGzra6unp+fDwQC6+vroVDotddeKy0traur6+zsDIfDQ0NDq6urt9xyS19fH8o8JSUlaIw0NjYyxwRN1sf7UiCuaheLOA+TVmAfuX2UdgRVUCsJt1j1GgBz+x5use797aPMUj0gZeLXtTxbnO3/prHeIojQA/k/3w9vbaz1mLX6QrKLYqyZTOyW3d3d5eXluH543yJnhRg8poeAHd+E1L9AEySgMdZEixUVFZiY0tJSl8sF9ocnSymN2+22WCw2my2Xy62srHg8HqRWobLR5gNjTTSgihIIyqamXHjBcpJ0n7wQKENWlOyvYpdFh0FQF5xQAFzcHyyyuEvQXeLxOM3JpCOwED+k3Q9NaQcGBs6dOxcOh9vb21HJkFbCWMm1tTXkFdPpNKi95OvR02hubp6amnI6nU1NTQ6HQ3jKIJ5ms1ktgpBhqa6uJmEoDDAJoRgxq9VK/5Tdu3fv2LGDhj5sq3Tk4anZ7fbR0VHU+nmCdrs9m83CncCaG43GXC6HcNXo6CgIL+lWgHLRhOFNFAF7enoofYSFxiY3NzfX398fi8VQHaiqqurp6Tl+/HhdXd3g4OC+ffuoc/nYxz7W399Pbqa3t9fj8dx77719fX0U1CwuLp46deqGG26oqqrq6+ujIjcSiUSj0VQqNT4+fuXKlbvvvttisUQiEYfDcc8995hMptbWVoIbAqnZ2dmDBw/C0NiCLs0GLEZZ09lWXhBCIRJAhkb8jG0aU32gqbE5m9mHou2dX2+y1PIcNYGpwbgLnk1vtTb7sP5s+rTnds6/2fulBw8e1A+EOpQFA5Pt39hmBZf63VVDCFFl/3CZVc8aX6yuri4QCKAfJBleDIf0L4f+SQhvNpsBkdEhQZwIuANjDa5HbofdOJ/PE7xDy2d9Ypg2Njag2VKNjWXEr8TC0vxUzZ+Ad2OCybxjgABzVExZI38juiVi9AFq0J+S1SIpKW5cWkeqMwxfGA05GgQTJ6ruBitWkvh79+597bXXaKQNJzISibCHAQQD35vN5ubmZqPRWFdX5/P5qMMmP1ZXV/e2t73t4YcfbmhoaG1tzWazAAhVVVUOh4MclCiZcRiNRsHZmQZihflfiqrZfWtqakhaOhwOp9PZ2tracO3g2nw+H2VK5eXljY2NFFU3NjYODg6S5JSkq9RfQJBHSBJ4R5QukGGgLNDtdh84cKC5uRnzTbVFOByemppaXFxElTAcDs/OzjJJduzYcerUKbPZDHoej8fPnz//4IMPvvbaa9RMDg0N1dfX79y5MxQKpVKpWCxWV1f3/PPPm0ymW2+9dWFh4fLlyxaLhdYBKP2eOXPG5XJ94AMfIMCiHNThcOzbt2/nzp1Op7OtrW1+ft7r9TKpVD6yJqqmIEUFRmT6qcYah2Z5eZnqXPRS9JZha9Oj7gR6ebjtRPbF27ODgqYKKlKw5uUtGett/q54hP9hxnrrH5PdWMODVmF7fct6kWQSFEylG6vERtVSqyx3wnxIvtKfkH9FOjUYDBoMBtxbHEmiYHwxCCHkLijAZXIjP00dgdRHSc8Lck1CxhRNPq4KCgGuNMkZsFTS3HweSjh9S6UROAiMNFsgJ6nRiJGmvfwid60ynNiupCyYF3hMvI+UD4aGVJL6sAX743q4YLXPhbo3sAdYLJbm5ubh4WF6isOcKSoqCgQCPT09LpcL13VpaSmXy83MzNDWa3V1lc5n0NgrKiruvffeH/3oR+9973uXlpaeeOIJeNBYcy6GXKIEE2wekrWX+J25BDRE/TGOPyWOgCecx263u91ueBoul6uzs9PtdttsNoIes9kMGlNWVkaJJqMHaUQyAYYqg8Pp8Pv9jY2NgCoAKTabbc+ePdXV1fv37+/q6iLPTD45l8udPHlyYmIiFArhBDCR0Mj2er3s32NjY/v27VtYWAiFQlarta6ubmpqCqHKoaGhrq4uEHCIE3a7/eLFi8PDww0NDbt3756ZmWlubl5YWIAy2NHRUVNTg96pz+eDgIQuIPsQ90V1rtPppFMo2j7SSFOjjKax1MKgJTOEs0K2WTpUafxfzWs9r0xlRuvpHKox2drkFf+swdX8yhtm7k2FP1XCafugjf769X50wUMzjJrzaA41rFT5WnL8NMGoGR0ZI8F/5UKl6GhrChoOqQhJy6VrGnqpwyQ5MWF0iqKQ1DiJsSYWw9kURjbBL04KuR35FafTKcse16m4uNjtdkM7A9GWGSkSYvphSSaTlIoZjcZ4PM6vZ7NZoBLpjS0+rMi9yxgKTCEhp/TnVedZ0ZaHgODk8XhTMuxbyPhJ8h1TVfAzXE9jY+PS0tKxY8ewqnTto5hzaWlpamrK5/M1NDRIE1gKYcbGxqampigRpB7k0KFDg4ODH/7whx9++GGz2dzU1ETh+OTkpNvtXl9fd7lcwFb8yzjD0RZXTs1Oy42ggovqG+45D46tSFB4uCj19fXz8/PUN83Ozp4/f76mpoZMQC6Xk1CMfY6H6PF6VlZX/H7/3NxcS0sLfM3p6Wk2YKADLDKMbLPZ/NxzzwFZACuJljr1U9FotLOz86WXXurp6ZmZmWlqahoeHn7uuec++tGPzszMrK2tJZPJTCbzV3/1V3/2Z3/2xBNPOByO8vLyqakpYr7HH3+8paXlyJEjFRUVbrcbKTG2w9XV1VAotLGx0dLSQgM2dGnwKwXRCoVCNpvN7XZfuXJFjaS3SFmpk5+DBD4rSACQt3poalX+9x0bP3t+/AA9g3uLL75V6Fn9LdWWalgo8hMF9bXVpiLkzK9i1gUrCbfArzW3odIV5H+J0+VSNHudhlvNZUF5BmegUzhSxfCKhP5JbRUJNwT+8THJ+1GvKE14c7kclQX0B1BHB3jEZrNJpwygW2EgqB3FJNMirdDR4BZ6tcgjcD2kxTBqUoDAdwU7FteYdJAGuNc80YITS+aByJ4J+rE1OdJutycSic0sNXkCLFQymVSxcoyjAH9zc3MTExOXLl0KhULUxZSXlweDwdraWhJ3a2trO3fuvHDhwo4dOx577LFAIEDrWIfD0dzc7HA4VB9fLLLwwdU6OnUKqbi2KrkuPgTmm923srISSweaQVf1ysrKhoYGt9tdWlpK/hOjw7OQHBSaU3TspoDe4XAsLCxwj9KMHJA6EolQ4tjX14eQCFeSy+WkpSEqrLCbMpnMbbfdBtZx+fLld73rXSdPniRPW1JS8vzzz//mb/7mSy+9BJTPTKME1OfzodNksVgAQ9LpNH7x+vr6xMREPp/3+XxcOU40+UC4p1ywy+UiAlMXdUEIQrNm0bDOZDJopRLYFZR72+zQ4K4F/dbNvlW8Cca99YEJUoMz4Z9c93c3u5Kty5I1FQ+a1+ohmSqVz675DL7I1aIYfUG5Oogqy13FmFhOKmlXLXXj+dHeDbkcs9lMEsZ27XA4HHa73Waz2e12UvPVyiHGGk9ZGPtMd+5NAmGwESw4CxtURHQXuVrwYg6MO9732toaCtSUaczPz4M144+oJa3isVJ9ZzQaDQYD1WILCwtMYshMXFsymQSckYHFZcPnUpEfDRCxzQmkybWK5qSmeZiaI6IyHklY/T4vh8PhoMJFcBI5OQ93dXU1kUjgOZaXl9PPjGFfXl6emJjIZDJGo/HQoUONjY3V1dX/8i//EgqFxsbGcrmcz+ebnJxE3hqSL8lDTZpeDETBmEONUSSNgQ8OlIG/CZKOI0x5IUA5fgBRFHgUqCsiR2xF0nSKMA4kHQXR9fV1cc/dbjcPdHR0lEVhtVpp0elwOHw+n9lshgGNqFM+n29vbz9//nxLS8vk5GRHR0c6nY7FYlVVVfX19QhGA6/19fV98pOffPrpp9fW1mj4sLi4CC2E80AgsVgsPHEeEK3pIpEIHjRMqrm5OfS1wQNprWu326moVB+9Rp9ANaxiBDgDmL7AwZr2clu46nroY+t5rn6reEtDWdBvFaBP1f8TvqDqCV13q9jiY8XXOyQjJUZS8ueamk9hCqj2+updsJNTE8EUJ/+AG0sJHwrCKuNS0uKaTQPoAydUzTnwGVYjXic5Clj6cDlEZVRsEGuGZcx+aLPZhGUMhUAGmg8DEVKmQSkzkhRQd6HZQhQlD4lcJPg1Au3AqTabjUks/gi/wm3CvGZlxmIxlpBAfmVlZYTwUvCmkp1ldgpApFe50gRBW8whAgjRdKXhgyoapcGpcPSkbVXBo7a2Nh6PY/SpRZKSToE4ieuRfDIYDA0NDbNvHtJPpKys7M4773z11VdHR0dvvfVWyGrIP/n9/t27d0NKQ1FIsrga6Eyfi1YDZxUSUXWNVaPPFOIumEvirWDmoAOJoA3aA3wX1jxUGb6INB2qm8XFxe3t7czn1dXVAwcOpFKphYWFtrY2m8125cqVoqKiK1euBAKBYDBYXV09MDAwMTHh8/mKi4uDwSDMkJ07d+KdDA0NHTlyxGg0ypyZnJz8xje+8Su/8ivf+ta3rly50tLSksvlcN5RNcG9SKfTVN+srq7CEgHNGx4enpmZCQaDPJ1sNosmMI4FkIvRaATmFtBPI7dfcMoRmGsgKSEjFLSnWycPrzvPiwohrvrvFoxBxUarlcyaTUK9fnlTD+/oU68cqqIydk8KrNQo8OpcKi0rKr7qtFWUVxQV/wwHXCX1cp3q+a/+4le+8pXl5WWz2YyxFqlZFhs2C06PRplQpFLohMRsxtbjrjLvgQ7xX6hS0dCNBSKUZYCGoQC+qoyL+t1gMEgpuUqABciurKxE11EKZ4xGY1VV1dzcnN1uZ5eTorW6ujoCybm5ORxD6YMpCS7cIlUyhnI1SkUw6NJLsKKigt4FY2NjojSiJ97rp50esbpuBxlGA7iALtTiFcpwib1TsYKC5Hos9dTUlBoUSzwBb0xwJzqSJJNJdFcqKysZ5HQ6bTabA4HAfffdd+LEieLiYlQ4cL2Bj3hwi4uLNHMIBoNsnPoCCr3npYfsNGqRmq4fbGbCQxciuaBblESJk6HupqwcHCI1Qy7JXvIi1INkMhmmDepIy8vLkUikoqKir69vfX19//7958+fP3Xq1FUo3OPp6Og4f/48/XDvuuuur33ta2g3vvvd73700UerqqoymQwIRkdHxzve8Y4f/OAHBoOhpqYGnyafz1NAgLSe1+uFtuTz+XAgEEgh422z2VpbWzc2NmZmZshzylBgyHw+H3sAaCGBrEbUX+0kgLQAmVgp15LoUPN01Pn8ixhr9dB/frMWHGLK5B2UICUhr55Q36FcXqvERP6F3sohhlVdMixAwYdVl5lxFi0juR2mGeZUzo9rdVW+hpBQ6HGg78JyZfNkF1VbpVHIwLIX/inwH+/gl0lcDEeChy09WTCLardyNfjl5GqNk6awSrOS6SkHTKlysISazbMhcKuoqGC5Go3GqakpzlZVVeVyuWBZZTKZzQI03EwWPF6Ypp2C5rEVXe+QiF79ov71Ly4konqI+uvc2NgIBoOhUEhcDJ6C+mjg1WKswRAAmpma4Kezs7MOh+NXf/VX/9f/+l/hcBgiTVlZ2dGjR5E6qq6uRpFjbW3N6XRCpZB6aPV+pRrzumim2pquoGdNFaKK5gEfMa/Eo5d0unxX1FTU8xPK4JLj39DmLZ1ODwwMYE8nJib6+vqo/6yoqPjhD3/o9Xrvvffen/zkJ+l0Op/PezyeVCpFe4fu7u7jx48bDIaJiQnUrEjVLC8vh0Khs2fP3n333U8++eTU1BSJUHSggM49Hk86nSb+i0Qiq6urHo+HlMPs7Czo+aVLlzweT01NDYItDC/AXVlZWSgUIv0u1Hix1Prco2x70gZXxl8vaq+ZZpu9r4ZQP98839icIqwJXiXBqIEB9e6LfpkUNNCaz6i5xM0OKgyki6PcvmiEFRUVSWHXT2HnkpISarSws2TGBYwWkAtWhoxLa2ureCusMRG1ETUfCeRVCgfqGfpKVrVKRSULS1CmqiDKrqVyQvGsr+0ZZcXFP9VNRUauuLi4pqbmzcbqlRUVV2FW1iE7LZvt1NQUntEWpkEkdxkQzsn2xnnUD1/XNdYkWwouj+1YajXoKfijsqkU/CFaTIVCIXlfYCu9FAGfgey4tLSUz+dZEmzY73rXuw4fPnzy5EkK/WmaU1tb29LSkk6nm5ub6dLi9XoJnig63yxwxiZufe/qDcpk0+M/mtUljjO5TclMqB6lWlauwXaJpVZXV5PJJN0JINhVVVW98MILTqdzcHDQ4XDQSri8vLympmZ8fDwcDns8ntXV1cnJSRCkTCZz7Nixd73rXUNDQ+Xl5f39/ffcc8/MzIwKKkYikYGBgRtvvPGFF17Yt2+f0WgcHh5OJBKIQO3atcvj8ZCP5bkg91hSUkILMcgq8Xh8YWGhpaWFbLA0ISMfTjE6zQqIgWgWrJknql8ix1tlE/8fO0reTEGr3qua/FCvWW0XKV/XvNYUxOubHsifBFYSGETNgoh8o1Ce1MtTa57lh67a9F/7tV97q6MMxQqJGVQf+eGFawe6SJrPq4Ze47moH+PK1HhTgxXIlmgwGHp6evD1KFxEfshms42OjobDYearSt1j/5fOEYJRgl3Ozs7C4oKiQCIOiBagkIbr+DVFRUVNTU2Uh6iMcp4HJzcajcS/mue6Bd9mM0utpn/VJCf/y2VjLvP5vIZxqNpZzZlJbFIZb7PZotGo5pKkNIZ32NioW1tYWMhkMogRyjX7fL7a2lqv17u6ujoxMeH3+0E5GEw64brdbmjOElbr4SD9AZKmvtAMjvpC1VWXPV58ZwaN5LO4NowY1A64E7zJb4k4u5Tp01qFMqWRkZHBwUEikp6enmQySXPhSCSipnOqq6tdLldlZWU8HqfqyuFwlJaW0hADvDubzdI1saKi4sSJE5B25ufnqYL58Ic/vLGxQQe+ZDI5OTkJ7MYEPnTokJgGdbYsLi5aLBY63iIb0tjYyK6AU7K4uMhMZltCKEbjD0l5lFowJY+sYJOpLfghm70vC1OMrKYBE4eKd3Goc0kmP88OH455q4GhNc64JCHlA/pliyyBJjwtLi5WW/dtxuAm7Sz3yKa4BfCitwY/D0dSaMLkMQFzuQ5cra0j1oKxkr5UsuB2rX4X3BmPGEoGatSU1dXV1dXX10uSU/YYIZNInYtIqdEJBU/QbDbX1dV5PB70iEFsiDzwsCRaIVfpcDhYgeptSqXl9rfDLcaHLJY0X1eLXATYUguU9JlxzTvs8zabbW1tDe6w+lcB6OXriKhIyRJ4pRSkrK+v9/T0eL1ehL+RGJ2ZmbFarXv37iVPgA+CWpMsy20OCykN8Df1W6ql3mLO8Ka0JtEMMkYBn4MpQX8ppL3BhYHUoF1GIpHh4WGz2XzXXXedPn26uLgYQzwwMMAVUg8lUJ7RaFxaWqK5cGNj48bGRj6fp5wqlUr19PScOXPmrrvuunz5ss1mY29zuVzxeJzIw2Aw0Ftg165d5eXl0Wh0aWkJvdmZmRnwa6p1pMMht8zspQ6AvcdkMsViMYfDsXv37tHR0UQiYbVaZ2dngbM2NjZCoVBDQ0MkEgEj4l8ZUr24qAY92MzVKDir9bGUIOmaHtPqoZ/GJI0070vSTwZEWtsUlNHnkGBLfkiNtEDqxQm4LmFR/RVyiRrwYPvdrq9e/PY/ql4TlWxzc3M2m03t6KPZCeVaNZCfvg3YZnhTwV/nYNBHRkb6+/vT6bT12oFUMbEzuSy5DDhbJMHQ0ScbBhfVZrPl83mn0xmLxfDKT58+vbGx4fV66+vrRU5EkryShGR1zc7OUn6mufhthocayF4zgFL8Io6DWp1IalGdYXoAXXM2OdCiExEr9ZPqjFR9GXnBrox/Ojs7e88996RSqZdffvnuu++emJiAptnT0yO8EbpbSZCoNvK47uCUlpZCTNLDyhyaBKPmfmXCANkRuqlehcFgoA0urIlwOFxUVATgW1FRASg8OjqayWTi8TiKBbQPT6VSr776KsYdZJxSVY/Hg+Qexl1Ym8XFxfF4nFReKpVyOp2rq6vj4+MtLS2lpaVNTU1QetLp9NGjR7///e9jK+knt7S0dOLEiYMHD66trb388svkhMj9TE9Pj4yMeDweCfm5NUgv7LJWq21l5eo1TE9PZzIZpK9qa2tHR0cXFhZEjrGysnJiYoILkz1MzVfJIGtq1rcoCNBzNuR9jTkm2tOQhjc7G4e+2E+zi0B9UfMNMpHULiiapDrvM1ff+CUFjBWHScZH80W9rYMzJtes+ev/LmMtV4Njy+/hkugDbY1pVte85t+CiIHmF9U/JZPJ06dPX7hwoby8vLW11efzMTWZsvTfU89GCgjuNv4OfGE4reRhYJ4i7mO1WpPJ5MzMzOTkJCabamYpAcADmp+fp2wXY61WGaiXfV0XWwMNb+gOuJJEM/LU0QVk3yJIV33hrZ8gRYwMhX7NqMZazkk2gvY6QgEqLi6+//77R0ZGent7XS7XuXPnZmZmjh49msvlkslkV1eX0+n0er2wJ+kmI6FYQYLU1gdrQ8MbkUWlaXuvWTMqB0kdIooe19fXh4aGFhcXf/SjH+FDQawm4ycAHcSPPXv2PPDAA1/96lexaBhEEYMEncB3xq0RHCmbzYqGIpVE0EAdDsdNN90UjUahba2trR06dIj28CUlJePj49BFXnnlFdT1kskkQQC4DYJckqRlBGDpvel1vtFbEmB6ZmYmlUqRy2lqakokEjTfII2WSqWEP77ZkhSt418k9a13ronAAG3UyLhI8T8kM6TuHwUdIwF8eeKwLCR+TSQS8kk9OURv+lXYR0ONU3+u4A1KbbMKZgprazvHG9mbtwpboy9KPEs0XV1dDV1JTW7qCZuSIdSYBv3HNL8oDhF72vr6+ksvvTQ8PFxfX+/xeKi1ITYH0ygrK0skEjRaRPAaAjWgZHl5eX19PXI8gUBgeHg4FosFAgEYisvLyzU1NbFYDDUJwMSzZ8+63e729nYe6uLiohDDxUlUtyIyNvo96boHpGDRlKBjKbVzwn9nQnNCv99PNE3QGo/HFxcXjUYj9ysa83JtsIVUmE+ob6puH49JUAt4ylgHZE8Q6HA6nXv27CEObW1tPXz48Kuvvur1esfGxpARX1tbO3z4MCLUIDmzs7M8Qe5On6/XPHd1fqpZa9WVFs9aLWUSN1y0FeHD4O1KJQLa/5cuXZqenr5w4UJXVxed130+HyPQ398PE06qWPE6kTtnPOUiqVAVJVX8L5LP8if8R8pnKJ2dnJxcW1urqamB6Yiiemdnp8VimZqaGhgYGBkZgR9ZVVV16tSp97///Y899lg8Hse3ADYFd9bQ43BKhOWlqsGwzeTz+Wg0CiUMth+egdSFyS4IC1aTdtNTIArq6qlQACEIc5ifIzpkrmazWWleo6cLF+lWk/B6mcDqLq7XDtS0EhS+pqbaW51mGkZ5QYBR40oXnMlCFuQpCK9hMyRE79v9nJ61GibLj21WZ6EvZ9dErNe1ZereiA+FEoLD4QgGgx6PJxAI0HYrk8mIri6LimwhGphS8pPL5UBaE4lEPp8fHR2l7ovmftI7Y2lpKRaL1dbWRqPRUCgUiURyuVxDQwM0YeqJCXLxaPgK+vpIOa+vr8sC2CZ4LZ2kkeOx2+0o5YPeCObOh3n2aOvMzs4iMx+LxbDdEqCpSxdzL9LYKrlYFgyxsN1uj0ajFCWzHtLptMFgiMViiN41NTUdOHDgxIkTQ0NDGKNYLJa7dsTjcSao2WzGtCGeJZNHal9lgW0G1mtcJw3dRaOgJvWiYlgxlMAFknvgi8PDwxcuXDh9+jR97pFMSqfTHo+H2CWXy/FAsfIyFevq6qqqqh5++OFLly5pmGdyF7Ozs/STpWSfMZe0pBR/it2PxWJPPfXU7bffnsvloIiFQiGLxTIwMHD58mWsajKZ3L17t9Fo7O/vf+9733vs2LHJyUm/3+92u3fs2OF2uwuuHY3jqcGXo9Gox+NJJpO40qwU5BXFBaELkjy+gjlhdRWr8Ij6o2DonBOCmYZ9wHphfDQt1VcVOEIP2amxmlyY6ohofHA9UqFPhPAOz30zmdafw9OVn2M+bP+L2zLWmryfVGpiQEWIQ/YN9Zo0GTN1sPRYmN6TkihGNm2QYqLOrq4ulG7QKY5EIpQUy0/YbLZEIrG2toZnjccNKQpzY7fbl5eXg8FgJpPhatfW1gjz8X0Q3EH0vaysLJ1OE5kSnOKkUAyJ58iaD4fDFRUVDoeDXgTbf5xCKYEGh6VGPp/fkg5M8hVpXWo0GiGQuVyuaDQKf1xTESCrSIgKAoOIEilxw9LSUiqVoqRTjDUCWEajsaenBzXOhx9+eHh4uKqqKpVKHTx48OzZs+qk5+JJvdLHi0PKu8njqZII+nki0aIw3DfLduBk8QgktiPCoHJyZGQkGo3SBRxD3NXVBS4xNzcXiURMJlNjYyMOL/w2wYJUGf73vOc9d955549+9KPXX39dLkAT0lZXV2ez2bKyMqfTyXiK8AjoE0lyKrmAsHp7e7u7uwcGBg4cOMA1x2Ixhoi7Ky8vn5mZKSoqqqmpOXv2bHd3N0UDu3btkp6Wmma413WGQAPQy11cXHS5XOl0mupNPlBaWorGLGUT6ndVE6Y+d3EmSK7qnxTjKdi0bIQ8MqJS+YwaORXrUkGac6rXpn5RDw9qSi71CUOV9FKQYaXGEFunWPUHEYaKvVz3+Dk9a4ZAMsWq53vd47ogl8YN1zwnlis4ncvlAjn1eDwrKytWq5UIVz4JlRU+o9T1SbVkJpNJpVI1NTXz8/Nmszkej0MKdDgcUFmLi4vz+TyixojBr6+vR6PRycnJ1tZWmg/APwFnrK6ufqPW6FrFOX2v1XvZgpknsxC7ib4+OvfoRrHT6LFvaaVYXl4uJFnkTPGv4efpN12x11KnJ5Y9k8lcunQJi6zWPnA2s9l84sSJX/7lXy4qKkJgDw+aShBSuPzrdrupeTGZTGx4S0tL6GKTOcCHEi6zilGqnpRgF+LfcbA7qm6UYDW5XC6RSGSz2QsXLoyMjKTT6aqqKhj3xLxlZWU7duzAqqJdV1VVlcvl0uk09I+ysjIQBhkxwaxfffXV1tbWEydOcBmaXhACUIK0iKCKqKtL/CSaeczSlZWVWCx25syZhYWFnTt3PvbYYyUlJe94xzt6e3vRaUA1m8dUW1s7PT199OhRCKxlZWVUnGvMtCYtpNpx1U4hYmO1WtPpNOdBI557QeqWaaYWhmg02jQuXUFtZPXapLWFuthhALOT6ZuEbby5iPRcEc3/6n1t9WFtBv9q3tckHvV3ocJK1z00do/bv24phhwFKg40sZJ+J5Gnjt4uWkv5fJ5SxoKPTZ03qorKZpZLk41Uufc8xfX19UAggMBIUVERDiyTHikoj8cDD4/lbTQaoV3TSrG1tXV+fr62tnZgYABGDgl3wLtkMklBAek7BEZsNhsfgFkcDAalWpTy3IWFBYARNSut39ULzg814SDMHhYMdSUAIFwDnfQWFhagxxqNRixURUUFKJAo1cGGJsukjq0G7ZWejUKQD4VC4XB4cXGxtrYWBxkdD0gOmUzmgx/84Dvf+c4/+ZM/EWoKg4NgtLC+HA4HAm/8HIgNRg2sVpXfwqVyOBwAytJ+TLooiO2orKxcW1uzWCzicrLBhEKh8fFxun3TpxUSIUxkroftRwIRmtJGIhEc/Lm5OTKNYCCAtvLIyBk2NDQcOnTo+9//figUIlslk1nmvwCyUtJNO3apF0PWBu0RaCokFfv7+3fs2NF77bDZbMvLywMDAwjycZHkQpHPrq2tBWzByBJ1aYwC9wUWJzMQHFZd0SwrBoctzel0ZrNZAYIx09xswcoRTYk/FDcZGb2nTzGO0WgUsQ75K09fo1JS8rPdFzXwi6Dzmjc1/6vZWraINlRoSxUL09+4fEYjZaW3nLLWNHLeBa2u5uKvuoBFv5hOq2yt28me/XyJY7F9skXL1CGUkL63sKeLi4vJbmWzWZHnRyoIZNlgMExPT4OgkWxEGgWrRIansrJyeHgY3gIgbyKR8Pl82PpsNjs4OIjzq7la6FkoLZCCU0kyoveveaiS7FIdB2r0ZRIwfUE/WX6YOVaFeJQ4aDwXTkvwIX6HSj8CBhFtI9Yw0DPVGYjnxeNxg8FgtVrRA9m/f//tt98+Ojp6/vx5FLjm5uZ8Pl8ul4P1cf78+fX1dSRDNzY2xsfHoaOJdor0A6TpFG3OuR4B+rl9AgsZlrKyMnoDra6u9vf3z8zMjI+Pz83NUQkFCMDTZNteXV31+/1S88LgSNLPbDbHYjHk94Q1sby8LEIL6rMrLy9Pp9OLi4sf//jH3/3ud/v9/kuXLgnsqCbWxDqzy1LSjUuB9qQUBGQymZWVFbY31FypSCSfmcvl3G53dXV1Pp+vq6urqKi4cuUK+xzSS6I7UVZWFgwG6RspswvjBR1eNQoa4qxazQ+4h8kD5oLNouqLCtakWigNR0CtFOMdFeCSZ82sVrEUPTlE7wJv/KyAlB7k2cJSv1Xjo7n4zc6wTesnm/dbolfL8VOxKE0IoB8jTbAjAjeawGqzm1F9Os0vFhxu9fkJT1ZMdklJCfQPSmNYSOXl5fTMrq2theckppxZS/1hZWUlymdSKBWPx6lqoS/U8PAw2HdZWRlQI7k+4neq72g6ozfWHHjcRLhUeHIXKuQnbri6WqQUhVQYLUvAVbhfKjaRFQVMkB64OLPYZWjj1N9LwR6/KwKtsgVirBcWFqCTC1RqMBii0Si5NXCe2tradDq9Z88eZE4xfBjTyspKt9vd398/OTkJatzR0TE0NGQ0Gqenp1955ZVLly5JGN7Q0DA/P2+z2err671eL00A4D4CMc/OzhItRSIRNsV8Pg/hoa+vb2ZmBnqD1WrNZDKQ/eFWlpSUTExMBAIBiMN0lpD2ZnivjFImk7FaralUCoXrVCrFZMaM6vVjwdwrKiqefvrp2tpaSNB6apcGvhRCiHrU1dUtLy+j9sVJZAwzmQydP3lSpKyXl5d37NixurpKyTh0GpH5XVtbQ1cvEAjMzMwINMS0kdaL6hVKj82Ci1dmYzKZZCEQpqgWh/m2Wc5c8H05rXxYbTUlYJcUBhdMgapxapFiN7b+kwaN0ZxTY98K0s+2Y9xVLGE79FwW+GZF7VsfP6dnrRLO1UvZ/hmu62Xr91jRi1FpOoRa6KASWvJJu90+ODgIG4+wDuY/kwYUmIVkMpn8fr/wFpDrW1hYiEajJpOJGY/HR7BZVlbW2NgIrXuziydOp/ZBKIwaaWaqPORmAXBl/bAeMNawTfL5POgYBXIgevj+clRXVwNGmUwmoe6pTXZUJpYKLtGogZNns1lshwTFjFUsFmtubn77299+8uRJujKurKwAj3i93pWVldOnT0uLLKvVOjY25vV6cWxHRkbOnTsn7bh6e3sR55yamiKZlslkysrKkNivqqqicoTs38jIyMsvv4ws0crKCrQHsKmVlRXUpS0WSyaTgei2uLgI6ZiJwcbJQ5SwAz8DBhFkauGlSCZQs+pMJlMikdi1a9d73/veZ599dnh4WAK7gskl9U0cT/rzgngYjcba2trJyUkB0ElQo3PZ3NwcCoVIlgBz9/f379y5M5PJjI+PQ/Q2GAzpdDqXyy0vL+NYrK2tud1uDWsCOW9ppqFXndY4YWr6ZGNjIxaLkaVg28OwkuQX8739tgP6g1OxeRQUctr6hBub22j1IL7c/lXJYinI0ttMsfK6Jl62Og3GLXvJ1jf7xne29q81SDTeh3QQdzgcLC0N8KThSGp+WGVoipyFJrOs2fdk45KqTTHcZWVlqVTKarXmcrm2tjYYHVTfFhcX+/1+8icsSFxFJBGgUoVCIY/HE4vFpMjNYDC43W54V/i5CPKVlpb29PQAmpMMKS0traurQ4EBIbTR0VHsPspqQnNmQHB29NWeXBIoCm1xaMkKcxx3nn7k1O8EAgGAAqPRKJKYxcXFKPhIExDWMLOEAZTlqooXwhtho4pEIuDC0skX15vijqeeeurXf/3X9+7d+8gjj3i9XuKDxsbG/v5+i8UiilqM9tzcXCgUwtywyKurq2GJwf+DLRe7dkxNTVGywSDMzs4ODg6eOnWKwAi0hJQpFGBRhaVlBIoZeNCUFOJ949iq1B1K88GR2LljsZiATkRgoCKq8hka34cPH77pppvGx8dfe+21srKyfD7PXTOSPDgRQdOYPzA0YeAgxJHNZrl9VpPJZBocHOzq6rLb7ePj44LhsFmSsO3t7V1aWorH44QjqVTqxIkT999/Py3YITjRGR00STTUNCxjjb3QR7riIDO96bRJHwOHw4HalMYx129XouCosQDiWqnvCAWF9Kz6+eI3fYuCFk3894IZL/k5fcW8Jr2vt4fqoSa39SZbj4MXPI/KeGYzVq+kIEb/Bk/mhhtuKHjSggCIXAezHD4GDAT2czwyTc2LSsqWMF/V0NmMJiGn0ugNAtVRPi4hHsXl9KDJ5XI2mw3+HDq/yGoLbQP9IwBBCg1QYyDNEolE8vl8IpGw2+24YygqwFUgv0fFBwQDzA0iSgB8qDPjuEn3VVX6jr9SJUHXMVkYHo8HYMfpdNbU1GQyGZyaWCxGtE6IgKXL5/PZbBZB1+rqajB3ng7dW+BjyckFddFs5tgvLnVwcJDFKQpE8qSqqqomJibY/15//fXR0VH2KqQnHA6HZomC70OHICPKuFEpWlpaihot2AUit/F4nBIPg8Fgt9snJibYUNl1EolEc3NzY2Mj/XmRVFQjEo0mnB6+U/+VCiC6SaysrGSzWeAUIUioCxWEIR6PHz9+nLYJqni8mmDUiNlrVpMUUtM3mYpHSZ+WlpY6nU7pxEQoQGOB5ubmeDzOr5Bu4TPBYHBqagr0j8dNECOsJBwsIZtLC3kALmIvOSSroVp2/BKr1Yp/vbCwgKqMlFPpEQA9rqIeauyCXcaLEuEtjdxH8TYyXpp4Uf3uFoiHOmFU1r+8o7E8aoG7xkLqT1vwOnlqaqGD3gbqv/sGK6vgDQjOpc9OkOkC2iM/s7Kyks/n1VSGek61GkJDC1F7ghS8bTGRGsYMrj0BhdVqjUajLpdrenqaqyK/RKBKvEyCkT9h8sbGxrDIAwMDnZ2d4+PjkPboQ8Y6oVfh8vIyEmg0jU4mk/wuFeqTk5MkEhFjwwnNZDKU7Qm7EaErGVWNmD3e+vz8PDURpaWl0Wh0YWEhlUrV19ejVYuaBGNCdA/CwwoHe6ERDKZcekJukbqRQ4hlmUyGOJ0FCc1gdnZ27969p0+fnp+fv/vuuw8cOPCXf/mXVqt1z5496MxxAWzYwKwQ9WD7kRHFsDY0NHBJZP+y2ezAwEBJSUkqlQoEAm63GwONRZB4CAEWCgunpqbC4TDDJVWX+pyKJNAKGmvE/0TMjyYbDodDJpUmVs3lcvfcc8+HP/zh73znOyMjI1yVhpopsbZc0tb1xPwEHccpa8rn82Drk5OTS0tLUH0Q+66srIxGo3v27Onr64MGA4hXUVFBv+bjx4/39PQA/hAq4coANfAVPWSh/1evPCf4Mu1mQOTgpwtdUg/Nb0djTz00zc41V7WxJcIgfAeVna3/QEEUQZ0hqjD3ZmOlp/S9JfBHfpQITJWI2Pq4ilFuAa7rqeD8C+Im/curq6vdbjfrTc6gbm7CnwUJldckvjQNkjV2Wfqlqru3SjgTiQ/iSpYcMp5ms1meAT/EJA4Gg8w5PkxOifQaNqu0tDSZTHo8HqI/kjkOhyOTyYyMjCQSifLy8traWrvdTj9G4gwm8ezsrM/nI6AjuBbLqCp+qfciDTjAxKE0ud1uKpIbGhrGx8fxgFZWVpqbm6kvNxgM8XicG2EbGBwcrKqqslqtoVBI+qXpGSCabkxcHugKCSshaQEalJWVdXR0XLx4Ea2fV1991e/3WywWapQ7OjpQI2JHATnBZEObwwHM5XIWi0UaTno8nnA4DL2SHuctLS3cuMlkOnny5MzMjNfr3djYQGcDmbpgMLi4uHjlyhWU/q1Wq/DDNBN1s3Ui03hxcVFIKaK3B39Rip54oBzr6+s1NTUWiyUQCDCRMpkM9Sxqcz9BqPXNDDXhJtAQcRKguclkcjqdc3NzdJ8hdwL7O5PJVFdXEzU2NTXNzc1dunQpnU6zgUUikbq6OnhQdXV1lIwh5QrhnamuHw0ZE9VCCTtIPinNr+ksyhbL9oaNVkOQn+PQU9k0z0t/bFZ9vhkiscWfeC17s+CWW3jKjNJmZO2tDwkB1QbEevKF5itXr5CiUlWGishdGJpwQpnQZP/pzoVPIVIvULLwYbFuCEHgW0lKh5sEjkC/BsYCQS4KnPK/asMBJjdXKzU4GBrOTGKKLM3U1JTf70deQ1q6MPX5NxQK1dTUIBaMWlNLSwvae1w5BR3T09MrKytE9xRA4rATS54/f55gs7a2tq6uzmKxsHsVFxd7vd6hoSGpnJTr12e6OYTmsbS0ZLVaKfgGH2htbWXVoV5fWloKa6WpqWl8fNzn81Ex7HA4wHDx0P1+/+XLl4uLi+vr68fHx0VpRF+urRZxUSAnuqmU77PXguyTtW+5dlDuMTc353Q6Ozs73W53KBQCTIDYQGdkqRJCz5ea9YqKimQy6XQ62eO5EmRFaTfs9/tnZmYIEWg6BRhls9nGx8ctFgsUPenWKPZaRfr0uRaJF+n3Bq2TQYPQmUgkWlpa7HY7JXx1dXWhUIhkpsvlGhgYWF1d/cEPfiD8URAtqZvX2BdWI4MGmqHnt4lDSqUfpBq2eQph5ufnWVnLy8u0Ua+urq6rq1tfX2fAiRKwztPT00yVTCaD84Hph16p8eDURlzqv3qqn7wmpI7H4+RU0HFmRahb5mY++3aMNQQYkhZk+MVL21Bg6C1icU3WV7xgPQdjC/j7uhesriD1fX1hjmYkSYeQuoDUT8SpEd9HtFZ1fMvq6+uxyyKwIswe/DK1iZnU11EdwAxjCjJZ8f6kyTf0MnGKpfJCem7Kbo9dxolmxWpyHVKnIJC3JHZAA3g8BoNhdXW1trZWJjp5cJHlZP3X1NSIFyPBARFQNBpFGzOfzweDwZWVFbw/WFbZbJaoAgYVReq9vb1nzpyxWq0GgyEQCNTU1Fy4cEEIyBKKbqFeRsEbmBLFOPF4vLS0NBAIgMNYLJYLFy7Qgy0ajcKKS6VS9Gryer0LCwvnzp3zeDyofW5sbDQ1NeF3I8oj+7nI1mhMGE8cBpsANTjCNpvt7Nmz1dXVNP9eWVmBMkx9Mwi7YAiJRAI03263h0Ih/EeHw4HVgLQu3o00uuQwGo1+vx9cG+nEsrKycDjM5dGre7M+OJr64ILGQkXqadS5sbERCATMZnMikWDC0HKMiA1MDJ2TX/mVX7nrrrsefvhhCCTy7KS3nrpPaOy1BIjyAR4BqVTJmVNfVl5eTkUiWywyZNC0QahAmbDp2Gu1ZD+dTsNbZ/GqHBhNY0+1DmszA6S+qXawwwGy2WzZbFZTqauZ29txPCUppW9Ooq/nfkuFIHpX+rpWuKBOiObguctIaiotNdiATAC5F+mdhl9osViI3dVvqUP6hpVnDgmwiE8nKoXSto7JjQsjLUelqkJIl+K+CcNXTUNLbTGRI8XZ6oxnhpFPkxS2QK4yavwKxloyfsgHz87OulyuyclJfDq73Z5KpdLpdDKZhByCL9/b2+vz+aBXG41GWF/8InrEWKWJiQnQavzc1dVVlGiQHMGms3myHrDgMzMzPp8PdrCGdLHZ9JUABYqh2Wx2OBwWiwWOGkpAbW1t0WiUAGJsbMxqtba1tc3NzSFFT0mF0WgMh8MGg2F4eJj5gX+9nfmN6UylUqqGJLJWAMQgG9XV1f/0T/+E8S0rK6NCEiJKIBBAyZ4u5iaTiT2+qqpKai/lYpaWlkh5SQxYVFTEyKdSqVgs5rt2IDBNgxhGY+slp/GeCjZhkuwQd8TU9fl8o6OjZWVlXq+XMcTpxikhJstfOzTuknBmqPHRoA0qdreZZ8d1Li4uSkUuC4TX6K0DT0mIhscj9hqeBugNA+7xeIQSw1PAWREFPk05ScG6D/3QqbMF5wkhGlXjoeAnOdTYQjWLIjHEC00+UJiv67qsuAbFLTjCGkj2LTGGN7PyEoIIn13N8GkO3sQkSuUE9wXMSEt7jQKffge9utjkmiQpLJ41UY9QkShbIPbhM2h0GAwGUALpvi4IOpgGsAYfwCiQARdvSxTIEP5nPmlQITUU0ugRE0dUV1dTQxwMBukGgP2yWq0qVl5aWrpr1y7cWPxBwBN+LhqNEpDOzs7iWePZ4U9Bh4KZRzhDSAjwBAQ5OTmZy+Xw18CCrkt9V4FshIbr6uoo3JCmRNIZr6KiAtVjn88HpACBzGKxUBxPczUo0kDGyWRyi6nJjoKCs4jpsMeAQc/OzrJDUxm8Y8eOd7/73d3d3Z/5zGfA9PGtSktLqXOJxWKoEvJXQDOIBMQ67Afs3+p0nJ2djcViiLrgVO7YseP06dMQSOA7bubEvdWDDaCkpIRBhvHm8/l4ZNSjw4LA5x0YGLhy5crg4KDGHBOSSq8/eV99rW98pTffuVyOsJWW07DmzWbzzMwMzWIky8cHKPHHXi8uLkajUYS3mIe4MlKLwNoR0EZDBNAUqmiuf2szh8YAwOD2B1+Pz2Ks1YKRrdNuxW8RKVZP+5YqQrb+af2IqfNTM1dBkkV3kDoG9nspAdmM+nw1xp2cnNQkoNR4RA2RpNe1UC850eLiotqCTL3uzQoUOVKplNlsZklQKj177ZDnVFDjWM4jAPTG+obh6mQ2YAXQAAHCi8Vi7DdAnMxXmuB1dXVduXIFcHZ+ft7pdNIThPgOrtjU1JTRaKRKkGBCFJ1oJ0q7EJpVo0FMgh7bKhdcEKfWDJdERgaDwev1QvOw2Wwul2tiYgL7JRTg2tpatOcrKipyuRztx5aXl1G5C4fDBC6QUjQ/pKegsi9WVlZeunRpfn7e4/Gsr6/j25pMprGxMdxPAouFhYVIJPKtb30LfRI80OrqanLLgUCAUntaKvMVSnjojE5ohZIMJDP1KcMOJrjJZrOhUMhgMJAMGBkZwTHh0QNzaah1AkdIFagmgcYiEcKZ2NDFxcXGxsZ9+/YdP358YGAAtxRZ87Kyskgk8pnPfObGG2/84he/KHA8uxeWUcwo2rnEheRXSNYVZFmANbNWoRizw+VyOfyV5eXlRCIhXdCkMyRRLz+EyQ4Gg+zN3Pjc3FwsFoOxDuxLmxJxmwSK0ZTVqaueSVKwp4nG7WWxMANxyPT9DNWnrGLlYj1xLMRF5bvEkXoAvfhN86JpQaU3QRpUqmBcq5a/aw65GBVBEkKEnEqtZNYnMFWFCSkYBmBUZWClnFhk1zQbzNWktsaeajA+9ee3KcqsyUiod675GJl3rhuhd1C8rU+uH3HCQJfLlcvlgDuo1qGvs8lkampqYpFvbGy4XC63211RUdHR0UHNC3gF9APqm6H3wcIGJJHfknbmJAQAc+CQSCdNXkip0vaBCCEGwPzF+yMwJ/MZDodN1469e/ciDkcSCRrsxMTE7OxsS0sLFJHFxUWYyAILah6BPHdecGu48DabzWAwsHfKJ7PZLLFbWVnZ/v37a2pqpNgSBBaCxEMPPfQXf/EXdXV18/PzMFVgWANqgRdhqfUjwLxPJpN0DgTIHhsbowKFIiMx62rHHFk/W/tNXKGsLvbUsrKykZERKkpeeeWV5eXl2traqqqqmpoag8EAQRA+Bvwftjf2dYnNCUdUFSdZ5BquqpgGKT6SvQcOj5qOVsEct9udz+czmQwioqTNkVcEZSKbDYpCoGwymdjhQEj5OQBPISRwSWqjBgnh9c2exD6KoRD9ftEV0UzpLR6H2B++KFormjBls0PFRjQGV1+cwr+SJ5C/ysxRCVryV2IaEGA126mHKeT8BdkmahtxcmwFARy17kTja79hrDXZ0s1KVNTL2j7qrzGvmj1HDnQhNMIuW5xToL2Noqsn9Hg8KJ2Ojo4Gg0HkiV0uF05KPp/3+Xw8ldnZ2enp6ZaWlunpafbtubk5t9vd0NBgNptdLpfJZBoeHqbE3GazUaUiP81mKPQYUKDV1VWgGLwtQfYFlpJloNkU1XEg3KZEuLKyMhQKVVRUBIPBZDJZX18PBtLU1BSLxUKhkAiIe71enP2pqSkIuTArIpEIWjwWi+XUqVNbDCOXhLdLxhj+CedRJwBTtqyszO12FxcXB4PBoaEhNifWMAotTzzxBJsNdh9hT6joi4uLgpjpxSrZXTweT1NTE81TeLI0zIxEIn6/3+l0QixBsk4uj7oSKZAp2DMJSglPEEiHS5qZmSGRC2tlbW1tfHzcarVS8YS29aOPPjo2NsaNiDyWdPOSh6h6kfq2G2LTeYfBLC0tFb5gf38/lUTCJ1O9tng8Lsaa5ugwVZaWlpCgYUcXiiqcK1A+4BpUtKSFhVhYla0rRlxjsgv2f5EaK5GA19RDqp6mxntVdaBYiZS5ytLmYwXR6qI3z6Y6ufrRLpjo07yj/1eoxnLx2E21XkTzQzIyLBm9sVYxDU3Uok9LFrSWpfv27dPzWjSf1kdAxds71BNqohhhseAcwV2THqZbGGt56pQsctt0uAD9cDqdmDwJAyGH+Hw+xEIRz3O5XEAEfNdgMFy8eNFut1+8eBFWgNPpxHmZnJyEIKyWdXEL0idJjDhmmvCNjtfiI2yG/eFvclq3203rQrZ0WIPipycSifHxcYnaIPmipwHlC7YZ3cpZqOlrh7rZ6Gcw3vHU1BS0MJvNRmSHb45Fk26N4EWhUMhsNqfT6cHBQfRJWPmgHLFYDHwfzh+IM5LNiCtttpDUsj3o1TTkHhkZsdlsyWSSu6PqT4wjB/G+CvbpG71DN+zo6JicnOTCsH3oVgMjgKeDp5H5+PrXv97Q0PDII48I2is/JLwR4Y+KTVQlQ/WcCpkSTGOStKFQqLi4OJ1OM4AExVLjSu5dVcTGO5ubm2MLxMAJZkWZLklUggDgDklNidXQT0h9o2SN06qxCeIGAuno9dP12XU1qoPHha1n3NgLJX9b9LPxuqZ68LqH2lEXNS72KtmxhEtDTYDGE5dBI8uirwKdm5ujGllqx/XGWuOPa0zcdvRVfkrI0zfY5nmIJPRmrvRmWI9acCGbs2bQpTpRVNU10cFmm4w6XdbX141G4/z8PMs7m80Gg8F0Ok02fHV1NRKJwDHgetDqBKNg5aOlR78uQOdIJCKz3+fzkYPSQEMS/PKwSenwUEHPVUBW32dSDT/xg8rLy9H5RPmISpNwOGyz2UA82JxqampmZmbq6+sRIcnn8+A2DQ0NBoPh5ptvXlpaojsJMCLqbmazGTVB9RrwoIH1M5nM8vJyY2Mj/jUfoPMZ7ioDTjdIq9V65MiRQ4cOnTp1Sh4xdY8rKysWiwUEvLa2Vm1WKQ9O5oy0TOR/FxcXvV4vWDOKK0tLSw6HA8VBBGYlOUlIROUk7lJ7e3s2m4VkAsUtlUrNz8+jHzI3N0eDtEuXLpEOhS1KNp90BUxzEavD800kEn6/n6pXm83W0NAgvCYCNckx2u12KeAWm6X3MRleGB1AGTabDbSaknqhXWnKsqWFFVMLhh+EP+lmSR6FaBLEHx+IE+KvoNqIGy5QkrjJ182HazwqsebU6AurR/NhqRjSh85S2cSo4veokejGz1a0631qFXdWv6hnlei1QVT3GVoLX1fFMPRwsXqonDz9XzU3q1pX1iDTTGoMee4SkWubD+i3VtUU6nutq683a3izzUO9vc1iAf0hgn8MaCaTWV9fT6fTzBJYumSBgEGLi4vdbjdbIh4KSvBlZWWxWAwttPHx8draWrZWp9NZWloaDofR1EfFiYyizCqNBAHl79IQFmVnyecUHGR5XVlZmUwmY7EYHZVYxhaLhSJJp9OJ54W2XFNTUzQavemmm1555RW3251Op3FDQNu/8IUvYDKamppaWlqoywfCttvt8Eb0lSPJZJLcpqClOGJqBUdRUZHP50PYLxaLfeMb39i7d68+fUr3A/EK+RM+rAiWMlZCG5VfgTaK9ZSZMD097ff7z5w5A++ls7OT7AIEZHFz4EfPz89Ho1FmBUJ9MvVRPBflP5HWY27QHpd7DIfD6FwDmudyuX/+53+m8Xl5efn09HRnZyfUI0m3yDiwc0jKTt9JXT3wEtBeHxoaEgMnXosmelAlmWQWadwgm81GbTANkiBNihQEYah00YWbKKQvwfc0JnULUFSPPIBQGY1GbkcFuwXH15wQlqS0phQmklDFN362EYGqBCCKfRo7qLGw8kJlrKmHVOGpEkab7Uyb3fjPcbB3giNJUYvaffinn5RXGiu+2c/zGX210havNTxzTW5UT4EUf3yz2xMpalkkGAhcbN7EukFHAz/FrSDUdV07IEVBAsGQsctNT09Ho1EacNChMR6PQyXGjeUapHeJ9L6hdoZKCo/Ho/ZFLhgWyTvz8/OJRCKZTC4tLVH4xxXi0RMHVFVVzczM1NXVJRKJw4cPHzt2rLa2lnoNl8tFStBms5FNfe211yYnJ8+cOQME39XVtbKy4vV6NbQtacxGT3Rp6SKUeVUSDG47Zg4dpRtuuOHd7373k08+qeKAlZWVKBwRm/MmDh1+Ex0jpfpffdacGaSYEkqLxTI9PW2xWJDnjkaj3d3ddJnBAC0sLIC2s2lRIw5AxA6K3j+thKnoAUlH4IU9A6KemPWurq76+vqxsTGfz/cHf/AHI9cOiC7JZHJ8fDwej9fX1zc1NbndbrPZTPwLBiKtBXnQ0LMYBz2eUF1dDdWyuLg4kUggki58L1wNtVMXM00QZDlUg55KpUTFKZ/Pj4+P19fXA3CJcgMLhGBOT7kTjSF1xsoXhSOsTmbNATjD3Wk4IaLlon4Xd0pK59RGNqLuWfqzihR6uVeNbdEQyTU6Hvo1KCJ0qp6w+nP61/KOxmXZptaH+KbMWwQG2GJRhdR8voAatWYc1cSIGs4XvBP9UG7tHfNCNk+VPKQZcc0hxVR4YUAEc3NzkKbFs2ZGyvxgWdI5sLq6emxszOl0RqNRkkjoLzOD7XZ7RUUFxbv0UUUnWm0SiKWAq4uhKSsrc7lcVqsV+YjFxUWZdvqIRH0Rj8fR4nG5XGjqI/pqMBhsNpvT6URiHx6hyWSampqCrdHY2BiLxWpqakKh0P79+3/84x/39PT09fW1t7dT7D46OhqPx2dmZpxOZ21tLStWQC1aNyAnDc4uDEKMtUpAFPNBWEOH4u7u7meffVZ9NOhakG5V+Z0kxAj3VKhNtTUi5AaT3Wq1+nw+cAyz2UwL8P7+/pGRkXA4XFlZGQwGeaaQGiGDirai0+n0+/0ejwcff3l5mWJ6FiRWAAQTujq+ttfrxWjSehi9AXZl6UextrYWDodpAH/DDTdg6yUSV3vs6pvsicFFmgqoPZvNSi9dKHoy81VjqsG+Cy5AuoZiAdENPnXqVH19vaRbiVpE4Fe1jKwmzfnl2Ayf3OwAmdRcv3jWqoEDixPBDXZxohYKC9bfBJ34itRGqWRiPbtJU/ov1yDHz+Ak61f/gyuicvU2IwW8paHY7OC+4Js1NDQgujI2NkY7ec2Z30gwFsS8N7ug6z6zrZOE6mCRy5I5ajabEfAlNNPvinIA1YlSM+oH1HTxgJENikQiGCYET3K5HB5WWVlZc3Mz0R8VBzAx7Hb7hQsXPB4Pfb9YnKR31X4LkrUQ7x5HErcRgwuWqtlm1XSNtBNFBqGzs/O+++7DUyZbSHqTrghsOXQ8uOGGGy5cuPDOd77z9OnTKIeUl5d7PJ5cLtfQ0AAI29zcfPHiRY/Hs7q66nQ66ZWF/qqq0YNrGY/HR0ZG1tfXnU4nLhhEZiG3iC6+3+/ni8Qf4XC4ubn52LFjuAaka6qqqtra2tbX18fGxlZWVvhKMBjkA1ARwApEXQv0A6PJOkylUmx1tbW1o6OjXN7s7OzBgwctFgtAU0dHR0NDw09+8pPy8nI83/r6+oaGBnBYi8VCHWk0Gs2+eYjCGepRUoHCFkUlEZWEwEeZTMblcj333HM2mw1jQXEpozQ/Px+JRCYnJzOZDOoFEB+BIIlLKEIhzJfFabVa8bUJ3ZaWlkKhENQOtkkN5VaFAkix4HNQdI44nxQVc+VQSpBboiup9B9g0kIaAaIVOpa0cJRZSqNL+V+pNFaTk1wqRp8pzR7ApZIllkM0grDC8iaTB4qONJPFwWTDW3uTYSXWXxOXq7ZFVblQ7bJ4eHpN17XVa3yPsp8SEMVqq4QCffbyuvREzedV28iwV1VV7dq1i+kKRof4IoAVA3v12sRYq1nBgj+29RVormabuw2TGNOwvr5eX1/v8XgSiQTXyiQoKL3IspcbZhCp4Eqn0/ATEIoizhVeKroW0Wh0ZWXl8uXLFRUVw8PD/EQ8Hg8Gg9lslm2gqqpqdHQUhokqVqc2n+XBC2wK6YfiUZnW+u4bIteLp1NeXk5Xw7GxMcgbTqcTggeyf26322AwkEwT3giuInpVNTU1POPm5uYzZ84cPHiwt7e3s7NTdFFoxI5qBMw8ZIwQO56amgqFQhaLhU6sCL1KsRxUa1AUSUaxkx09evQDH/jA3NxcX1+foH5kaE+dOjU5OckGw7AIXQxUDgfcZDKhXsJqREkcgjkp4uPHjyN/kUql9u3bZ7fbl5aWfD7f0tLSwMDAhQsX8IWbmpoQtn3ttddCoRA5AxGJFUUEcs5Yc7gu8uCwv4uLi+i9RKPR3bt3f+QjH3nhhRcmJiY8Hs+OHTtuvvnm0tJS3HNuh1KshYWF6enpZDJps9msVitwLfrg5Dmx2qoan0AuNOihXyUzFiaosI9E7IyUNWMlUvr4GaqqpdCHMNPcNUGnbIrUTGEZVQ16UrjsoGRoQL2lIYZ0HGdiS5sbNiTpWI3Wgl4bWjozqBKymlZhkmTDkHElq2+239W40ts/NFCGvhHi2uo1aLe4cEpwC1/2uleymfEk1VReXr57924a0WFMwFHJqFVWVrLwf8az3vrUb+lStvMt1bOj+2dxcXFbW5vL5Uqn0zMzM0KDk/5s6plB5YWyimZFeXl5IBBANG52dhYaHE4HkALy016v12Aw1NfX5/N5kkvMb9bJsWPHPB7P6OioxWJxOBxCh5KGtgL8scND1MWGIghF03EJikW4VgA7DuZ9XV3dLbfc0tbWhpR+SUlJOBxGLw0lIwoj4/G42+1OJBIHDhx4+eWXDx06dPbs2d27dw8NDa2vr6dSKfiFJ06cQO2Exlq4eHBC8Jdh0akeEHgL1wNkJr2lhT+AO1ZbW6smdrD4mUxmZmZmampKgCnw9zNnzjC8DQ0NIjPGF4Wfx09MTk4mEgm6pcA8Edx2YmLC5/MFAgESEgsLC01NTadPn4bkY7VaHQ5HR0fHgQMHvF7vzMzMmTNnNjY2/H6/2WyGdY6gmJhLTAO7FGZa0h7MwPn5eTz3eDx+5MiRPXv2LC4ucneowezcufPAgQMmkymZTELZhvQNaD41NUXihOJ4gGMRL2OvIopnNCDeoGe7trYGA52pKMkuLDVGkPiP1Ah0JjGUrAUWgoipYY4R0kIFjE2dxCa1lzINWFP8ELsy2n6k6LkM6WWqkrUlosKF4moFflTZUyr/VW13oKpjSos7oRuqW1fRtUMNTwsyrDbzJlXSi9ZYr61hqfUowtbmWMNmVt8p2JNBdmvihvb2dr/fz1QR1mBlZaXZbKbthiBpb60H4zY1tPTf0tybbIzEvNTFoeBRWVnpcrkoYFXrRDTPiRFXaxDwAbPZ7NTUVHt7u/Qmh2TKPKOnXCQSwVCOj4/X1NTQaxEAkcpdXFqDwTAwMEC8SeJI8un8i3wSCKY4lRJIEvBqAi7hAoPMdHd319XV2Wy24eHh/v5+KHpwHqjxgcHCCABenz59uqWlZWBg4Oabb7506VJLS0sqlbLZbAjsud3uHTt2PPnkk/v27btw4UJ9fT3479raWjweN5vNoOHS5pg4i0pf9Cggk6m5GqPRSDW2ZteknZjT6bztttvOnTsnnZvD4XAymUTfldQZ46Y2C2YTymQyuVxO5SMvLS2RZkFx8PDhwz09PdPT05cuXVpbW7tw4cLk5OSNN95IOf7i4mJLS4vJZHrmmWfon1JRUdHQ0EDvcCBsZpRaqYR/gHcpRooZQsoxmUyix33q1KlkMnn27NloNLq+vt7e3j45OTk8PFxXV9fe3m4ymXp7e8GXcYLy+Xwymezr64vFYt3d3QARAo6zY+FKz83N0SiHFmVYbSSxQA80OC+mEItM7pTnJU0SOLhHaQjA84X+yHwTHjGxDt9qbm5mTMSxUHOAQtqRaS9EOg3gSSky04AIZjPSakHqmwDEQqGRKaoX2fgPOX4+a7b9o2BaUhgplIC43e5gMEjeW1hYEP8DgUBzczNt3t6QENHvSwWl0/U7jL6ORnOgQKY+M3ULAswiHbdz504SVk6nE04ojDTgP7LtOIZMCNaYmkYQwARUenx83GQykR3yer0QjXEE4HiAU1MGRvSaSqXouDg5OckvItpHIbvkfETFGJ4NK620tNRsNhPPEtTncjl+RSrlmMQorAIF1tfXkz2LRCKIf4L/zM7OAm7AxFheXm5raxseHiZubWlpsVgsIyMjJBUPHjz4+uuvE3VGIpE9e/b09vbu3bs3FovV1dUlrx2ICxJhIUYoZB4q/o1G48zMDBuYCkfOz8/DoiGwlZpDDtAhwISampqBgQGbzYaDjBWYn5/H0LPTJBIJEAOok+JPCZ2roqLC5/MxDYqKipqamjY2NoaHh7lsWMxzc3MDAwNEx0tLSzjj2He45EQngkRDS1B7ErH+sc7SEp6/4k7SVOyee+45dOjQ5z//efqhEJayvCORCHN4586dMzMz4GZ4mk6nM5fLTUxM0GTL5/OBxqAKKbxmMpbs4lBTCCaIDsvLywH0uCqoL7g1TqcTajzrRRq2rayseDwe/HRCBNjK0nUIMCQej7tcLrvdHggEmJaSORDPV5Qv2TJFWVNjazTMXcZZGl/pGb16holqSXgKQuUU+7O8vMzyLHtTLlyfD9vMChV0hwnK9U0X9UiApkxkazaXaos19fqqACybEKXaBoPB7/e3tbVJWx+KoagMIDnh8XgCgUB5efnExMTVbPDevXsLDoHGt9+CxbIZ4U8Im5Icl+CF2e/z+SwWi8lkAlbHN8nlckTc7e3tXq/Xbrc3Nze3XTvMZrPP53M6ndD+VZqOxWKh4630LCe/Tx/C1dVVCseRxs9dO9Buhg1GVEg9N5edSCSoXHc6nVwbK4T5JKQIPCBVdd5kMvFsoBngYrMgxWvw+/179uwBbMHYwRcuLS2dnp7GylDTBYg5NjZmMpkmJydHR0fX1taOHz9eV1f3ox/9qLq6+ty5cyaTaWZmJh6PWywW7qu5uZlqabKCyE3gnYmmqzw1AQdlQUpNo8ShCEVpAjoM35UrVxAQHx4ejsfj6LuiIyqdwsn3kj0TOFVWTmlpqc/na2lpqa+vt9vtJHVLSkqy2Wwul8O4426wgTF/rFYrVUhWq5VdGf90bm6OO5Lclz4Tw2v6VQp8L/0JyZf80R/9kd/vP3/+PHxKbtZisXi9Xk04ZbPZzGaz2+0OBAKiO7GxsZHJZOLxOEQ6EGcSR2z2rIXFxcW5uTluE5qp1K/LlqlGHhIfsBuRHqTnBuk4rg1XVJAfsBemot1ur6+vl62CR5BKpYhpILZSaAo2mEwmCYmYDHpoQrVrEkXp7WBBQEA9ZM9Qk3giHFrxJufqF/eFNUiFvK+hq28nbXjdH1JPBSKEHSA9TrSKTcjlcpB3F+avJvDAeNna2d2j0WjZdXcYzZtbqILoOTQFuZDAMUxxngfJ/YWFBSJEvuLz+VwuV21tLWaLzDK7NxG0ek5QRchzaBQ4HA7KHILB4OTkJOJNmFp0qJnHgPcWi2ViYgITPDg42NzcTGq+oaEBaQVpTaApbVIfhoTbYpQpFKaSHm+6sbGRCHpmZgYZYlouCS9QM4doEojnTnuUtrY2etTSfoHz22y2TCbjcDj6+voaGxufeuqpjo4OKjn7+/vp8sUn+VedoKx28Bnp3qQ+MpiI4oRqKpgqKysPHjy4e/fuqampvr4+OpfTqhhwU1pTooAhUhWgOhaLRRhBYqewIHQ/wfgSBjKz6YDudrsvXbqEQ20wGBAxh0fBk91CsJQjlUpJUktlZCKT8uMf/5haecw9PeOB2riYioqKhYUFi8UC1IOjevjw4Wg0GolEpqen0+k03XXZ+Ovq6vx+v2DNuI1IypBLYM6IH6A+IBU2ZI5hmuVO19fXwRKFxiASfXhzuJOyxAh9mMPJZJKgEAgVFJtWnJDWCWvUBokyVsQi1+Ujb22j3wC1r00ofV0ilJWNQgVlW/SP3SZpeAtXerPC7OueWV9SL/41D7eqqioYDDY0NFAUwuRPpVIUphWXFJuupr2uCp8JLGaz2Xbv3v2GZ73ZiGuuUg8ca65MBZexyIFrBykXDqBbQYHp1JfJZMbGxuA5yRYXjUbJ9hAagBiQs85kMqrRIVBiSjERscjcP3OUDBUMMDwp+H/xeFxqPcEx1tfXk8kkWqN2ux3mAGtDCh9Ell6CO9YDkt943JTtEbd2d3ffcsstVC2PjY3RsxzfnzEMh8OAtrB3YYbhpXZ1dU1NTUGVNRqNnZ2dAwMDH/nIR1577bVbb7313LlzDQ0NUvRFetNisdB1gabsCFoVTLbIHFLL7cSpoQC9uroaQF+V8pEGAuw6VMb7/X6Hw9Hf38+Q4tZJ7Gk2m2tqahobGxsaGkA8KHom4zc/P5/JZKTtfTgcJg2IueQiEWHA9q2srHR3dwsDV50SQjbYrL84bjgXJrw6liiYz759+7785S9Ho1E2JJpRLC4uivnG9jkcDmIO7D4/TbwYDAYlcAHOjsfj0mBatIWJOQSRkOSbqnAieK7g/qLTxLQkqSjKULD30KSVLk7EeVynw+F4Y3CKrqZbuWX0dtiKMOKcHF0XUVaQKjtR5lG9yIJYhz7zpv9MyTWChlo4LtWtpG1KN9EE39pwbXboZ4VG4q7gt36+31KZY0iN9/T0QONh0SWTSSwbtAJYUih9Mu2hCVw1cPLDTBd8demeJ+last4cUIKYUqq2gPSmk1QGryU9jUFZWVmhhDqRSAhrBz6vpD44M0gfi4R6Nv7XarWqbmAul6M5AHW6NHwpLi7GcUM8gU5UrApGDdtNBfbMzAyrbmxsjC4tgKRoTZjNZvgeOCB4zSxyAV6BO1gtROhOp7O7u3vnzp10tg2FQgsLC/F4XMQtyewh8gBNSji2yWSSEhgRdhgZGQEncbvd1dXVTz/9NG3G2HJIItXW1mLWp6enMdANDQ1Xrlxxu90ej4fKFw3eJz0ipDyXFungM0hLb2xsQG8AgueL6IkXFxcPDw83NTX9wR/8wTe/+c2XXnrp+PHjfKCrq4tyeZrsLCwsgMkSa7OD4muL1qhMJBLLYmhsNpvX60X4v7y8HKbzHXfcwWbPOKjESnGONEUN6kRVCwWZFQhINTQ0/Pmf/3kymRTfikyyFIXBcoFrMTExgZljnqfTabfbvba25vf7ecT19fVTU1OpVIqeEpyQClVh0dJwB18YT18tWVSvn2tWBYOkqSlIBZ4QSocSBQKwzM7OYiAmJyetVisNMOXJyqBh32mDGYlEgLDwuMVz19cBqm7p1pZa75++cYPXtDPV/RV7zV6+8aZfSKih7wijMabbtKfqNUjnRn35oiaqUJnaBTOo6vnFByJ69vl8nZ2d0oxpdXUVp8TpdFZWVsbjceIeyhdxcSjguKqc3NLSIs6UCFwRugqfQUPGkP8VXwCXUxWKxZyBlqpdC9gJJAxUwS/ZsYUwD1yFOcNxQA9PziDjCLiDugXYNOaP3CB3i08tLHScFIfD4Xa7XS4Xyr+ZTIYmXnTUBbuwWq1DQ0PUNFK8w8gIx16iaX69ubm5paWFVrD0nw6Hw/F43OPxyBgWF/0U+Eb4BmJAKnk13Ja49eLFiy0tLRcvXsRfNhgMbrd7YWHhlltu+bd/+7cPfehDjzzyyK5du5577jmqXfr7+81mM9g96A1isP39/WDoKv9JJRgRLsi98Cea1usnN1NWFDCYrGfOnOnp6fnKV76SzWYdDscdd9xRX1+PNyqycJLxY2Kwt/GvXrGBR282m+Eyszti1y5fvtzT0zM2NtbS0lJTUxOLxQYGBthvNIHndZcrH+Y6S0pK4LecOXNGNORkP1NTkfT2xUSS0hTXhJIrct0IHnR2dqLQRNxjtVqHh4cnJycl66sq0EvCTbU+170RVf5BNW0CrYgpZ93RvTOTyaBspbGbkiQkyaF2j3yrhz6S09Rh6HkKBcHu1Wtb41v9af0J3+qhUfDXozFbjwy4MyGU0WhsuHZg0EAd6auJv0WBAuV7eMNSAEwKofS9732vlA/xbOSjstWoEoIigahyhwVWk9IDkUyi7AKiGxguZCaj0Ui7UpUyKRROtgEwBKR+yUNSmMulq/WjuORQ+kktygagcu/okUhSnpR3SUnJ8PAwVQlUMeRyORQ2uFocQIJuvshtSvqUUJTVsrq6msvlyLbH4/HJyclYLLa6ugqKYrPZGJCr/sv6T/tBoJM5NzeXTCap1oEYR+XFwsJCXV1dLpcLBoOlpaXgnrFYrLOz8+zZsw0NDfF4vLm5GfC9pKSkpqbmzJkzZrMZBwp8k0Be9Cv005qdUnB5whej0YgUlJ46qrYmWllZGRkZuXjxYk9Pz/79+8+dO7f72kG5M1nHxcXFbDYr9ErJN0qzTf0CQDW7vb29q6treHiYYSF1WV5efvToUbr/nTlzZnJyUnAACTk18klbhM/8Fr4qEgW1tbX/8A//wJ1KW2fVbVTlI1QqG1VIsVgM3VrgBTQRvV5vMBik23I4HE6n01AC0uk0C0Tke1gsaptEOdRMiXovqlNF3EaUIItLOqhJh3KcCTLhUsAlYRYcErwfsu5qI4WCahsau6z2n9Q71/KOjK3sEPJ5vBbcvtJr4ICev7zFUbSNo2DfdPUK9WfTvHNdFSPpjtTQ0NDc3AzKytOh1+jKyorZbCZ2AXiA3CWtEH8Khqjiwlg3DDfeuDqamD+5N0kHiVQYzw/jJdMXiIDpiB9NtpDaBK5pY31j49ockGeGZafOG2owNWbix8lOy4UhjopLTutx6hgJ/KEQUJ/GSKGbAQcZXgG0p0gkQlnH4OAgvirw+vT0NOEqzj64Hi6MuNXSu72qqoriPerXAXD0guVSlsJmg63BWcOzXllZ6enpGR8fb25uzmQybW1t09PT1DQD146Ojt52222PPvro3r17T548CbwTDodbWlqKi4tbWlomJydRSXY4HGazmSYpKtogUwqoAf6ZNFczGAxOpzMSieinoGS38Q2z2ez6+nogENi/f//58+fJNOTzecTyjUYj9pqwhgJF1fqonffkgNadz+djsRgQE5SApaWlZDL5wx/+cNeuXUyhoqIiWgcImiyn1TeHk9o/FaMUwsORI0d+6Zd+6dFHH5WaBXFH1BZioCJ40yiaSrTHBk+MmEgkvF7vxsYG2fJ4PN7Y2DgyMuL1el0uVyKRwL6rrGQONXmzRdMpMS4qk0eK1NUoSoBv1v/8/Hw4HG5oaCAA13SBEJVnqRT/uZ3TgpetsaRbG1apYCx689jmxWjw6K11X7dD7dhizDf7Ft6bwWAIBAIdHR3UoJIyJZ8EvRVmAbOXScULKkVYOFfh7H379kmLVZhzqVQKATD8QdxyKlaBRzCF6XRaPil5EqhawvoSv0mtQSKbDzGZLPzatWom7DvfpQUfrDu+ju4lWvuQqPjfNwoxr8GFmEsSR7BrgT4TicTq6qrNZsO/ox/j1NQUFmRsbMxisQwODlL7x/pHE5UQHi8DGBfgXlpjiGkAY2FDampq4hYAZCkLTKfT0r8DGI4JhC1wuVzoedIERCoUGPOZmRmYCTMzM8lkcmxsjG4sXV1dzzzzTHFx8fj4OKBtcXExjD38Vog0q6uroVCI7U0jxq3aa4wX2eempiY0SKXXl3xM6owhocN9TCaTCNAsLS3V1NSMjY2dP3+eJwgtaWpqihukSE9thKhxZ9TENZkDWhKTYqLEgy0tk8lQ+wpdEvRGIiq1Sk3OzycZc3X9U/OZTCY/+MEPdnV1RSIRqCYEJaJOJ+MAoZDwC6CAH0J+Fic9c+2grYEEc0SW+AQsJSYhVp5oT9TD1a6SKiKs6Y8KssFfcTskHytJCCaeoBzsK8TdUABV9X32bOllrDq8GoOl+tTim4s7r/mihiitT0WqLjmHiALOXesMp0p2FPSgCzKg5ZI02RrV31dRGr0frXe0BT1Xb0dVf5WcM83hOjs7yaLxsLLZLHOgtLQUY822LZK5nIEeqvF4HEXJMhR7cbY13QrAtjc2Nurr69n8mYiIUuIRgGmopHpVxVE/juLsYGfFwYSEJLCggNfFxcWIS1Cx5na7q6qqEPegFSFpw0QigfkGmMdm8aSljjafz/v9/rq6upmZmdra2h07dkCuQg3H7XZXVlb29fUBv6ChTMpxaWkpHo+TKMMRlu1HnTQC90u9HFgVsx8ncd++fdFoNBQK8ZxkBAAryO+L+iDVdE1NTeFwuLq6OpvN0nqmsrLS5/OlUimfzzc6OhoIBIDaFxcX6VQAPr64uBgMBqenp81mM70IuOyCWiuaqcn1g7rqPyZF26RGJRx+5ZVXTCbTF7/4xdtvv/1Tn/oUmxm1+FBfit7KIVQHXgjgG4lEuAs0K8D6Vc14zRJVlzeNbwS4UNtrIa504cKFaDT61FNPYUBpFwnMpSbSpZBBEGEOQiipcYdOSq/6ioqK1tZWICBCRqlbK1jRV7BbtPy1oJOo8kZU1rC6GLngaDRKMFFRUeFyuUpKSsh1S5k783Cb5IctwAe9W1DQs966WasUy5Rd87K3rsVTR0leq5RT9dBg6JvdprqLiDPB09ffERMPRKGsrCwYDLa3t8vFQNbEz5COV0K+wExzVdhYUkc0Fi/dtWuXvsGPWhnIE1VVgKmLlcgCt2ILwp8K7fEO8wNgjuSkuBL4EZRygDmQh8lms1wDusYQkAGIpXDLYrHEYjEcH3Lr9KaDUkI3bogWdru9r68vl8utrq6OjY3R8lwaGBqNRqvVmk6nk8kkHWyrq6szmYzQIbB6DKXaCwZzEAwGaexC6ICRisfjZAWi0SiVEaJUSWYSQHZ6ehpnDU/H4XDkcrmurq5EInHrrbfG4/E9e/awFd94440/+tGP7rvvvhdffPG2226LRqME+Dx7QJ4LFy6QeADSxefauqUp97i2tpbP58PhsNvt1vQrEfIWMzWZTIKGr6+vGwwGq9VKUjedTldXV+dyuWw2SxihUUCW5b3ZMnY6nS6XC5IvY8X0g+QHRMM5edziS6o6n5o7VZ1N9U8Eke973/sOHz785S9/GcBNNgmn0+nxeNiW+DyehJCr5DzsbYB+RHXSbhXY2mazHTt2DK4evhVEUpXFr7HF6phoeHJyCBcAcrqGEaB3DLHU4lII4Y+NXH5dxTNVL1L/L06M5rLVL6qEPPUnNKrIGj+dnY8hrbpGEsP332z+qFGIZgA3u371ULnIBXcgzZ9Ub1p9LYQ0t9vd2dkpxA2MBlEvqTiAeDZ1vGwVrCfFbTQaEQv6mf5mes8ax1kSIOQc9G74ZrQV+au6uzIzcKXxLIS3L98V3QYqwfDjRG2Hg+anYOvJZBI5UEisGO7V1dWamhqca7Jz1JuRoWptbSXJ43A4sAWQYWlygTAp6tgsKmqdhQmrxt3qwxZ/kB0OkRNuRO05LZWQ4jnykCTW5rSo6LESpqam5ufnqWOkIzDqt6xPChdFTS2RSIA2ejyeiYkJyQyz4av+jv7ZSQ/yfD6vSXPheXHx4EXQHzHWTU1NZWVl3//+91999dWvfvWruVzud3/3d8kQqAnA66ZlOGRwpGk6/FOwuLq6uv379x87doyZjVCBfu7pp6jkHjUJao/H84lPfKKysrK9vf3kyZNgzdQQAbWrZ8OzZtNS30chWupdWXu4I+vr66FQCM4SnjWTX0ga24Ri9U089HeqRgwqN0ZaquNwZDIZbAd1OjD59I+7oLCEeJf6nJt6qO+rdvC6HDvOf/X6N96A0deu+ZQaAfQtDr2RlThDf1ViAK9rygqeX8MNI8S3WCz41BxYA7ZGeqsieoVoHbNFhLxViBUn7GqrjX379m0UOvgO/D5YgaT4RBV3i4ekuWFN5rS4uBhFBbhZuLeYQnxbcUZQC3S73Xa73efzofSEp0NGjvpaeM3gIdlslp4vzMJYLObz+Wh7GolE2PEikQhg0OrqaiqVikajcGPRIfN4PJDeMNyIipSUlKCyxtyVDC8LQGr/KPz1er0Qb9ktq6qq7HY7kAudoljkyICQmKKaZm5ubnx8fG1tzWazERNFIhGHw3Hp0qWbb775/PnzNpuNZpUej+fs2bN33nnn3//93x88ePCFF17w+/3I13GpZrM5HA53dXXFYrHq6moQf2nSzOPQQ8byjKjXJ/skrU+YDCKN5vP5Ll++nMvlGOodO3b4fD6z2UztMqJUY2Nj4XBYdPs2W1QFJxKKRexw/Ho2myX9QPIZUSQ0FuLxuCj+MLlVT0eYsOoLPGWWGd6o1WqdnJx87bXXqDg1mUxms9nj8RDiYGHFsUCQhB0RZVQaRFDwLWQh1qFoyxQVFY2MjFDGKXKs4kypmEDB+mya9eAlqKsMa0tWELdd3G0V75azSYssECrWOIUw3Be0BMkkFeRyqGxoPQSsioHoC2cEi1fPJt3W5beuOl5rbxTWl16bfsSgBUFnvZu82U5Q8PNS9apKYwqUL/8rh+Yzaq4F1lZ7eztMG54FKRysBG4QrjQ8EB4f84RQW7gMQsv5aQVjwRuTZjOCdiE4sNnnVWOtGU35E1PZfO1gKgOZ8bRAwPE4qqqqEolEf3//lStXpqamKEYg6yi8C0T7sIDUClIkSQRNJA73A5YSPWIA3AH7qT05f/58ZWUltYVS1GO1WhHtJMZnWqt8A0F44caweTqdTnJTojlFdMLqhXwmLZNx04RdNzExIT1hSUWCMuVyOWRUe3p66BJAI5vGxsZsNtvZ2aleDyTFRCJBjo6eZHAoNeW8BbmivANDnI1TYh2SDdLtZXx8nJiuurq6u7sbPCGRSFRVVb388svPP//87/7u7/7mb/7mmTNnRAVpm2612gSkqKjI4/GIlwG9Z3Fx8cYbb2xtbY3H49FoFLaZKjUuLzSlgPJa1VCsqKjYv3+/yWT65je/yVM2m800n62rq6uqqgqHw6RqJSqiFI0cI9s5quI41xR84jpJQQ3isdlsFlwOb0PPLtiCwCutq1XZE+ndMz8/b7PZBITR85dlY1BxS0FXyTeaTCYAB6YoS1LFB/Rpt4LWXJMUVb9ScB/SpA15jR8jwXfltXHmajWGXt0h9DIjmxlo9TJ4rGotq54gqN6+ep1q10fanzJteF4oCefzedY1W7VGX5ArZ5ChL8OilmYdV1Vnr7tm1PhR1ME3aylU8Iv69/l5zpnJZKLRKBp7sjjLy8uz2SxVPZgAtlbqzv1+/y233II3jbX1+XyQT8hbwjZnMfh8PshnkgsNhUJerxc6uYgu9fT00KbLaDSOj4/TY1Q6tlDHyHItCBaxtSwuLqIiIrqmIK0YOG7NaDQS05Ha1R9U5aFG3dDQMDEx0d3dPTExAXJdW1sbjUY9Hk9jY+OpU6cOHTr0+OOPf+ADH/jKV75yyy239PX1UWheUVHR1dU1OTkJ2UDyrpqns4XRBGXOZDLUxUjKjliBdvJEWpWVlbt27bJarVeuXKGReTKZpLkU7ccOHDgwMjIiDqPq3W9xDcKJhEqRTCbn5+dh5oB4PPXUU21tbfhfUBJF5EgtKtGQJmWQJSuztLT0q7/6q5/85CdffPFFYCV4HT09PaLCwUSC7CTyOixvIHuXywVPC3VW1AIkTw6VRUNBK3io7mrBxCNzSfpaycMCIldVn/Tf1a9cnkgqlSKVSs6TGlHRLWGExVptZoU1XZg1y0S+ou9YqAp9aFgi7KMAmwuKxK6KPm0GTWyWsZSwQ10FUrqp7u7y+YKtzVXzqv5KSUkJkvGC9NKXDplv6f/H5CHjTWWJVCPyJyECEOm+oWe9hVVVKdhquZr6metSFDUHPGvitaGhoZmZGUQwXC4XvAUUJMrLy2dmZlZWVvB8SUhOTEwgA49cKoYPrzyTydTW1mYyGbgB1EAy4WDRVVZWIliDSDTFO2hRLiws9PX1NTU1Ue5Js4/i4mKn0zkyMoJttVgsIONqbkr2/Dc4iNf2AynOFAABC4LvjPmGX6gfHDYPsRF0bZ+YmCgvL//7v//7zs7OJ554orq6+pVXXgmFQjfccMM///M/BwKBRx99FO20jY2NQCBAf6/x8XFqmhGzBhTTd0fdDN3D98zn82oRI5eHcYzFYrDNGhsbbTYbjL1kMklykiaKf/EXf+FyuT72sY9997vf/bM/+7Nz585h7jfr+K4eIkBRUVHhcDgwKE6nc2ZmhppGulbyJ4pUAWTI0MoFF7xTQR6AAuLx+IkTJzgzAcTc3NylS5d27979+uuvA1b09PQgFVJRUQGiTQEBveRbW1svXrwo0olFRUVUneAfiHLINosqNzvEr9S8jyOFOLt8siC2KzJSYno4IdrK1DTgRhAaSligogqbGV8+o1b8izlWyYhbe3iaKlZMVfm1NShVo2/J5hT8sDo4Ut2mbu3qXavvX33n2n/SlFmTWV1aWjp79uy+ffv4AA2A+AnRU2QFSRM4Amvh8spj0hA03sgPbHGfIj2OTvwWDJutB0hliW5sbExMTJw5c4aC5pqamvX19Z07d6IpQyFGWVnZ2NiYy+UaHR2lghwFDxqY0qBAOlgXFxdT9wGxEd4FYkwqg5tqRsR/pZx6enp6fX2dwmUajA4ODuLBlZeXB4NBqh7S6TQqGZo2ndKC2mw27969Gylq+oWj2oN0ibQyAommny9ayUByUFxEmYH61JmZmYaGhnw+v2fPHuGqBwIBsKmuri7pgtjR0XH27FmLxXL69Gngi2g02tLSEolE2KgEOCa6lNoQDTCCi0HSkjbhzD9ukF2koqIim80SoBQVFR08eLC0tPTs2bPYU7wDdoV8Pj8/P08d4/vf//7BwcFEIkEiRSqJNptCpJSRG8TYoRlrsVii0ShGHxlutkbCeSl1k9ZQKuFa4xZxTnaXp5566sUXX+RB8MnFxcWZmZkDBw709vZCNKJzI/mfPXv2nDp1iv91OByxWCyXyxE5MY3ZPwhQXC4Xgya96PRrR1X+k8KWgtUcmpbh8nXUw6VEXmNEGAq15YoAXBhohCd37twZDAYpNINEL4ZbE/hr7IBKCFGlQvgif1VhCmEDa0AMPRtaWqhUXSvUln54mx0aF1gaqukPySiqJEUh1KpDp3Fei4uuulxi0OVOedyJROLrX//6Bz7wgUAgkEwmLRYLD5c8EKlymQkC36sK48x/3pHumgXaeqkHTWwBT6WQZjNfTHNsBrqhhtrX11ddXd3T0+N2u2ERwsqAJmW1WuF1OByO1tZWj8fjdDqtVqvdbqc0k5thCCDMJxIJq9VKO675+XnKOtWqHC4gEokg/H/8+PG1tbXR0VHYJvT/npmZoWOh3W6HyDw5OclDgo1HjlGVOpExYQ9oaGhovXago+S9dkixz8LCguDUpCWJjAicZ2dnR0dHpasI/X83NjZaW1uPHTt20003nTt37vDhw+FwmGxeKBQ6dOjQM88809nZeeXKlaampqWlJUYvnU57PJ4rV65QSC0FxOCSAqRKekTamLI86FBFHAedWdIJKAVOT08nEol9+/b5/X7C/1AoBByvLjNqPc6fP49d++hHP3r77befPHlSyoK2YBPjo5GBoXcPIBhXhZgis0L6E8pyQj1OqPqqJyjXBkNjYWHhT//0T9/73vc++eSTiKwy8owGtzM3NxcOhyHq8UVqlNLptMgAoHyUTCZnZmYYJTGanId/AbXJLedyOXWNaLoLqqG3GhDIC02Csa6uDnqrRvleXXqqnyuer7Sv9fv9NTU1VqvVbDajkCOUJw0JT0VsJS+nWj0VzdDn68Rb10w/9Xmp0LOAjZXX0hJbsEFUioQ6pJoMjZ5SXJAWoWGIa1jnGpheqAcLCwu9147Tp093dXWBU5PCpeIJjWjRqJIMmZBipeekJCffGJAtlooouWyTK7P9gyvz+XzsMB6Pp6ury+PxwM+NxWKTk5P4levr66+99loymTxz5kwulxscHEylUlRAYN1EsAmcqK6uDtVKBIOqlaOsrMzr9dIjY319/fDhw6jikbT1eDxQRMjXz8zM0C8G60zdI1tXwTvC0cvlcrRKuXz5cl9fH/wBODoOhyMQCNB+BRoJax7COCpLUAPJ6eEbojFCTEC/m6GhobW1tYmJiaGhoaKiIqpgamtrV1ZWamtr19bWamtraSm7sLBADRE7BJr9qmqP5Jr5V1q3LC8vR6NRrlN6+1LMzVAzPsD6k5OT09PT0WgUZou6ekV/J5fLTU5OvvDCC2VlV4XDGhoaYKHSkF4dxoL5HOYrDi95PzbpsrIyRpvUrkjcqMl6jXaViqUCCmFkKysra2pq9O2jLBYLcqN48U1NTfRppBTWbrc3NDS89NJLV65cSSaTiUQCcoXaxohDBdNFTEbPzOFNjX3ZzlFdXe1yuXDtxeTpizXUr4gLDAzi9/t37tzpdDpZPiS48K83S9DxoFnLoq8pzEW1oznOAfu9eI7yDl8saItULLGsrIzELGQkOTS9HNU9T91dxPqrNrrgZNNMSP3kVD+guQbE18bHxysrK6PR6O/93u8RIkN1pT8RI8bnNdwSLDXLUE0gv7GsVDaI/jCbzTwtWNyCWV/33jTblAwNi9/j8cAGoYJDSNwUI+D9BQIBur7W1tb6/X7czLq6OsJM1qckH+hfjso+fV5g0cmgkKDABlFYePHixYqKCsqgV1dXp6enGVOCdJvNFo1GqRcHvEYRlISAjKMIWrKkKViHzrG0tDQ8PFxTUwOeAJKOm0ZsASRChxpCzqmpqXg8DqBP0JpIJHbu3Hn69Onbbrvt+PHjt9xyy+TkJCIhGxsb7e3tzzzzzF133fXv//7ve/fuHRoaEsGAhoYGCucikQigijC4of0y6Sm0IwnJPs9r7h3BQhiNMEABZInmGhsby8vLJycnYSDAhBPirTx6kq4rKyvRaPTUqVNDQ0PveMc7PvOZz6yurl65ckU6C2+2NvCs5+fn2cvZEmZnZ+FsLC4uEklweWJWRHRfgDuN6ZfuEGtra83NzXDDAS6knp4Lo+q1pqZmdXUVHd2FhQW/34/gicvloh8xtT+0NJMCVPENeabMtGg0imSNXLCkOvWOnubiVSujLjHUJ6anp0WAQfUK1UBQU/TBd/mWw+Ggj4zNZkMjmzmgaQip2izidG5T7V4PFMmBH8P78kXVqd/s6WOsYQ2sXSve4QGRAtVnArePZReszNwidac/v2Y3lapRnDDRlVtdXX3ppZd6enrosIz1YHMC5ROaMhacc4oqrBBA3whE9uzZs9mtkmwBuAB81CiZbX9QVJBeRKXptSEN7mhIODY2Bld3fX39pZdeslgsJ0+erKqqOnHiRF1d3csvv0yXALw86AGJRMLlcoFnMWtJAZWXl0Nx46fNZnNraytgscvlqqurY5N3OBySmxoeHq6srEyn0+j/IeqNbB4EEmlJxVQDJxXyjc/nA2Dy+/1FRUVjY2ONjY0UQUgzMOF+yWYjMBkZVCSoEKeen58PBAKzs7OkWCcnJ+vq6tLp9M6dO9Pp9Pr6usvlikQiFKBPTU11dHSMj493d3efOnWKuvyamhq2dIi9Gxsboh8tPaTZz0gnIkgyMTGRy+VQry8pKfF6vWQ8SkpK4vF4Lpdzu93wUmgLi2MrwgMaVSBBHrLZ7PDw8JEjR4LB4NLS0okTJ1jPYjVwo1TfjSkXCAQ4JzsNuvg4uTU1NejHylxXOw+ogk0qisqYVFVVffrTn25vb//a174mxT6yYABVLRaL0+m02Wxs/ywEdtnp6ena2lqDwVBfX09bCfY8fFW5DHQymSd2u53ObWyfQorVU1ZU12+zlSVaerW1taFQiEy7hkkiNeiqNLbqcpLMYJbSaRuSL7YYmmZpSWlF5RudyVi5lZWVYHcwu6T9rvA7pSAZZTRxcVSeguTnxTioOr2CNXOpi9f2dYozCvL2CkYAGssmT0S87K2J2Gqxq+ZZ6B1tBsdisQQCAXjrODEvvviiy+VqbW3NZDKymUHxJLBWO2HSjEUaWcgmpzXWmoOUF5IuqBqpF6qZN/rCioJwj/SbIG9gsVgoN4DXwhV3dXWRVqqvr29ra2tsbGQC+f1+l8tFnof6b+hxSKfiR+OwoMGWTCZTqRQ11lCVp6amQqEQ3t/58+eBDnHMmTHogaTTaZPJNDo6ygXDdoRJIm2WJH8lHf8oZmGHq6ioAMMlwy5CKOBuGBc0oQKBAP33qCibmJioq6tbWFgwmUz5fB5x1D179oyMjHR2dsbj8QMHDiSTyb1794L/NDU1nThx4t577/3e97535MiR/v7+QCBALY/f7x8eHl5bW4vFYlRqsLA1nYvV1xhuiHfQ4yjs9Hq9QoIEeGlvb2cR4hFIpKYaGs10EqnCixcvvvzyyy6X63Of+9y+fftOnDjBOhRJXvXCJPGCQBXmcmxsjCdCdExNkyaFKJ6OpoZLfNjZ2dk777zzU5/6VHl5+aOPPqqf2MALCwsLk5OTIyMjNE3HcaERjMFgCIVC7HMWi2V6elr6mfF1Cb1po2EwGBYWFkZGRij7FIgGI6VZ/AVdIjXGl/ALshNduMSt5lDVQiTXpxlhcRKRM5RsLaoGcLFW196AwjDTIjSmuTbVo1cbkpGLE19YzfeohxpYaGiCxW9WL6sN2jVGc/vA0Wb8kIIfVjdRdY7pdRqEGIObiIfBUzh27NjS0tLu3bvZGinmIt5SKeEUwQJtq6byjSe+hWfNFmowGPDCVLZZQbLH1q/VeyMQwLPmJj0ez1XW97UmgT6f78SJE3Nzc6FQqLS09JVXXllaWjp16pTRaPzJT36C6MTq6qrVaoWLTecqkdpBQY2eMoFAAI8b4Nhut1NZh4IXzbBhZUSj0fX19d7eXiwCeCW1JCS16E5CfkDABNUtwmen02NpaWk4HJ6dnbXb7XissnJoUADpjS2HvBzefTQaBbpyu93StyEcDtNNsaGh4fTp0z6fb3h4eHx83GAwHD9+vKam5tKlS6WlpZTwdHV1jYyMtLW1Xbp0CSwbQiGcNo30jB6AW1paIpqWKk3aL1AilMlkzGZzR0cHBFL6IYDzis9FfCdzV+YxU4hHjGrKDTfcYLVaz5w5Q4pYkAFxf9gDQPylhoi6U4IPsK/19XWPxwPOowrAbm2sUe/z+XwvvfRSf39/wYWKQOPS0tLIyAhuLJRKsgI7d+6knTEF/dJvU3roiGye6B0C30tyVcVetw5YC3qOuPAOhyMej2uYFTIImqhfv08LTwO0k7jT5XKVl5dTgHf1xpdX6ORCFa7omEtJJ+OpSeWJp4z4ibqLawAf1XwLWqJasZKfFfOTz+grazTvF7zlzT5f0L7pjbW+JlO+JVELD93v91dXV6OZd+Xacdttt+FTWywWHDu1CpQpRwQsORgxHW8Ya/Va5dECDGH7xOLob0b/zhavZfmxS5eXlycSCbBgSXlZLBac5YaGBsmlms1mr9fr9/vpzYzCKuWn9OagSQeRMtw14GYaDojeKXrKOGVYdlEUQb/JarWmUqnq6uqxsTE+wx5If1WyW5IvUpvgQGhBxclgMMzNzc3MzCDmt76+TpJ9eXkZz5rqG+SlUD6CZUitNrjE4cOHL126tH///rGxscOHD4+Ojra0tBQVFXV3dxOJl5eXx+Pxrq6uK1eu3HnnnefPn7fb7RcuXCCMZceigSRoL8ZaQ9RTHw1oCaRDEYVAbLe9vT0QCGATXS4XmVIaFLzB2H/zoDpLndMc4gKDYk9PT584caK3t/fjH//4b/zGb1y+fDkUCqmfl2UpvczR9V9YWKBmFbOLaCKigyoZVhNxy9KSW/N6vTfeeGM+n3/ssceAtiWZI9eAU88WYrVaUUQAxo3H44FAgPzKpUuX4LaCI0nTJjHWgO8UHFPJJYW4KkCxxaFZm5LTpkunKgCp96z1qKu6m6rKXCjfMtrQQ6XvmkAT5GMQd5W0jTQ4lQ1A0gYYIDWZob8X9RCvWcPkWX/zUYrE6893bGHE9PZX0/Vcc+V6Y63hw9ACpqWlhfUYDoefeeaZG2+8kSZKeHUa8U65U/AlSf5fnU67d+9WeeyiZgAaBbHMYDAYjUaYEgV1f/RcloKblehOcB3418XFxezV+XzeYDCMjY0tLi4ODw8XFRUNDw9XVVUNDg7Ozc0NDQ2trKwMDAxIqY8o1c3MzKytrYVCIaPRiKtVWlpK9Qrmgxthc0bvwmw2A8iOj4+bTKbp6WkCwEgkYrfb2c1wK5LJpIwPjiGQnIoAymuHwwH+YLVap6amQE4MBoPH40HLAgzd7/eDTNHgDmMUiUQeffRR4gDpZIY3V1NTMz4+fvvtt7/wwgtvf/vbf/CDH9x+++39/f3sAQMDAzfccAOtv4g5duzY8frrr3d0dPT29uKBEuSm02lhd6qH8G1xWhEIFKkKoIDm5ub+/n6RPE2lUhMTE2VlZWazmQ0JMgxCo7IlqHZBdHEBfyn5SaVS+/btgxnCNqM2h1NbjvEUEokElEpgIk5YUVGBspVgoPw0/6vOSa4KaZG//Mu/vO+++1555ZWLFy9qPFzB391ut4DXfr9/amoKqA1XZmZmZu/evePj4z6fj5CoqqoKRim+pMSmFN8SdiCcIGpossGIpVN50AU9Zbk7/FzBJ/XQgUB2+kFQvWCWJP7Kzp07u7u7kagEoCcKJLwmfwNUIkqBGBfhp7MDqcUyajJA9eWFIqkeKoVUva+1N6FtDQyyWU15QSqexi6pV6V3vTXPpSAPRP8n9bWYIJ/PRwlVPp9/7bXX6uvr6+rqYKAK9MFGyGaPHJukYd7gqu/bt08jegJPw2q1AqkIbgIDDA6DhpSuD1tUHpWQe0QBVXYV6c3MLiRsOWQ9yCISaNvtdojYbOxYUmmPUF1d7fF4GBEgDqpCRPZFbYLO1GxsbKTbOpUX1MtA7zt9+rTVakXznvo9XE6HwyGZIk1pEzX0NTU1ZrMZCxsKhebm5mALrKys0MIRrlhVVRWAO/bFbrcPDg5+97vfldbaS0tL9fX1kUiksbGRlqzQDIQhBKuEsADDF41Gb7zxxqeffvrIkSO9vb0tLS3JZNLv99OmE0BfbV+peXbcy/LyciwWE3iHD5BXRE4a8WjqMsxmM22GqFKJRCL0fgSPq6ys1BQjqIuKdU6p9/Dw8Llz5yoqKv7bf/tvt9122/PPPw/5VLqXAgISRVHbTbqJ22FCUi4oar1itTWwD/1rmEJHjhxZXl6+cOHC5cuXJRlIsh6gHGahlGjTtBA81+VyBYPB8fHx9fV1v98Phy+bzSaTSenIJT0QpGUSDmkikeAzUmSosacFuQp6Yy2Ov7A19GZIE7bL/idPhHdYFPwLl3EL06ahXasfkNvRe77qCfW+nXpspnFa9CaNZzOenz6G2Ow8W6PbmvI9PX9G/YzmXoSoo+aHmcCBQKChoQEw8Omnny4pKWltbaVajTALhq66pakUzKumk6IY0h1Wq9XhcMBKhmHN/q8K5GNw0caTloniKYNsEKGrRHf+lWp6oWQiME0JAyEk4S3qB/yVPinoTly8eNHhcIAboj5BeQLfRS+fkm6bzYbTKs8AEb54PA7c0d/fn8vlLl++TLvCZDJJ/cjq6io6D4iBhEIhuM9ihtRkiOpZl5eXezweEBW0/bDvZWVlVqu1oaGhpKQkk8k4HA4a6bJ6A4HA+WsHNgiHEWmuxcXFZDLpdrsvX768c+fOvr6+9vb2CxcudHZ2hkKhqqqqVCp15coVg8FAu9je3t5AIICOdk1NDVVCPG+sai6XE9MjdkHlga2srBCmyK2JODr5XsrxCVDYzKCHI2EoypCikqxCexqXR/RFs9lsPB6nf5vBYJiamqIzkaoWK6q+tGiR82P1wG0lhSW7qR5ewAFcWlry+/07duz45je/ee7cOZnVHGQ+2P45D7xpUROsq6u79957DQbD8PBwQ0MD7Ca73T40NIRbjQIUjioHVwtdBB6IwAV69FZvXAoyH5iW8E80TVj0iK240qql5k9ERbhBpBMRB9YoUKtmWv1XjYEknyYt9wraR41915hUjdCYvC59M4chQiWaL26RbNQXXm72WjNb9JwceV3QWKt2XHP9OB9Q9YuLi0+ePDk2NtbR0SHqjDxQjf6fui++IWILyiZeiQAumpkEsYHKK81j0Dw5PRKn30jxQLnKsrIyu92O9ceDW1paIuUIEk3iLhgM0q6XeBl5FKTCeIrAqUaj0W63h8Nhdcjcbrff78ch7ejoaGpqymazFouluLi4rq6ONmAzMzNlZWWXL19ubW0FjqytrUWWc3Fx0W63wxXRP1fAO0gdNpsNrhI3CDswl8tduHDBYrHgm7P5VVdXv/rqq8lkEhSIqt+qqqq6urp8Pn/jjTeCnjc2NhoMhiNHjtTX1/f09FD2EovFUqkUqOvu3bvpn1BVVXX27NmjR48+99xzO3bsmJ6e7ujoiEQiBoMBhw7+jMbVwq5hULDmqmR5Npttbm7u6OiA/AdLBLOFha2rq4OFIt3or1y5Qg5Q3+hAM24Eifl8/vTp05OTk263+3d+53cOHz78b//2b3/3d38nWgqSe8QKSEEBgwwETCJBrYxQJ57Q+KAb/dEf/dHy8vK3vvUtMhCqRSNwpBMN2A5JWuCdsrIyl8tFIoS6WaTDA4GAxWKZmJggUwqHSiodYG1CPRKW92a0mW0eYDUoDAviLHehLk/Jzqn1pZpuEtJYw2w2J5PJ7u5uEsXbr1je/gf0F6l5v+AJS960XGqbLskxaEbyFxnY7bCTNc6+vLnFF3H+1tbWuru76+rqXnrppdOnT1++fPmzn/1sfX09YQ1TUVOsIHJRpYcPHxYxciY9htjtdgNXSUQMc1D0CWVDgFiqJt/V7YghJvnDtGDj5SQIfSBdFovFKGXOZrMOhwP2LlE8cQSNCIDASLvTcICFIRZzdnYWmnY6naYBNj9N89ZsNltaWjowMDA0NLS8vHzmzBmI2PihmANA2JmZmerq6nA4TG0kYbvb7SYgIHrg3lmKNputo6MDwmlRURGKdzU1Na2trTj+kKuoNVhZWenv7x8cHMR88A5kcJ/PV1xcnMvlnE4nVBz+JaAhx8X+gZxQJpPx+/1ms/ngwYPBYBA4nhxRUVHR0NAQTDI0ZUTHRzPzeMQ4jxhH2MHEQHT2AVgnluczcGne9ra3tbe3I5zd2dlpMBj6+vpwHjUFGkwJTR9I6egmSfBMJlNXV3fmzBnp3IrT5/f7ISlDvaC7JlV2uLHI2IrsiRAKRSoklUqVll5Nqvf09MzPz586dQpRNNFdgb9YUlLS1NREM3KYo6FQCENmMpm6uroMBsPp06ehYdHME1Im0gVMDFwQOifAxjEajUR+oqOv7ihq68itfcA3/KxrfVe5TfyqkpISkDHNGTRUCtVxE9hHIuCqqirS7AB6Gl++4GvNOyLlsVkqS2VNqB/jUCEp1buvvAZeSSNjNDCEEqqactkg8UHRZNZfg+Z/9S62CFhrNlcBP1XoQ2pbVEleNS7RFBC1tLQ4HI6JiYknn3zS5/M1NTUJ4FxwJ7t6/jvuuIM7Aa6iXpGyQDikmmo3xkvelH6DmkMa5krpIxOUmSTGmloGycPw3erq6kAgQMlALpeDBgD+gNguVt7j8dBbAD4Ja8lms6EXAfgO7wLDStEKZzCZTDU1NTxLpOlKS0tHRkY41fj4OB2SwNaTyaRUglqtVtJTokUr/1ZXV3d0dNC2cXZ2NhKJrKysuN1uREQpXaP1TCKReOWVVxgW4A6poYfOiCMPu4t2ugCy7J1NTU0ul4t0VnV19blz56gMQr+QwKK1tRU8gYlC7p6NqmAtAA8augJ4iCxpnktVVdXExAQF9Nw7qerFxUUyZsC1ZWVlZ8+eJWThf9XfKuh3iB1hOxm4dkxMTPzt3/7tfffdd+rUKXoNQ6yG1gJKxjjjahiNRqfTSc5WCmTUYFH0o5nnPp/vscceW1xcjMfj4N0yMpg8j8cTiURSqdT+/ftnZmZ2796NH7O0tIRgenFx8fDwMIxG+uRSgy7GSHAeCQRpniDOjbpZ6juY6J24gm/yQ+xPmFqNjVBBT5HpUE0tE5s32fZw8Twej2aTkNpOFRlQYRA1ZaUHIvThtWZTEaukv+viN2vKpNQWqFA/kYQyKHkRkaDZpssvZ1MHebOmsuqNyP+qGkoq3qh+hfKZ2tra4eHhl19+ubm5mQIL5oykoFU8pLS9vZ1C4dnZWRqWg+0mk0mU8lnnKMNJRlg61IlLojmENSWvVfxR5FzhcsFlpsKNZth+vz+TyZhMJjQ6KENYWFgIh8MiEYnTTfE6/ixrEqjX6XRGo1FSc4uLi1hA2rHDfCQ5Oz09bTKZjEYjeCuqCFxqJBIB56UkhAeGkwvzj9UlmgDgFdLyCv1l8EqasvPkXnrppaGhIVJwEHcikQg3jicr3FWpfOPulpeXrVZrS0sLCpYjIyPId0xPT8NKZFpUV1fX1tYODAyAUONW0MAJM1dwqXALRDOihycmb3l52efzwUCXGm48stnZ2VgsVlxcvHv37suXLyeTyVgsxgajYSnoX6uTXnyThYUFOjqzrWIKgfJFKIpSHZmH7FtOp5OAQBT91aIMpm5FRcUHP/jBhx566Hvf+97IyEgmk6msrFxcXIRWzPVAFmppaaENsclkcjgcly9f5inE4/Hh4eF4PB4MBpEHoXLEbrePjo6KV6sx1tlsdm1tjYVGBexm8fI2jbVYBEFaxN/i2Wk6u2sU71RoWIw1JwFdQREMD0mSzKqunubk6gmlbFU1YRo3XOPp6z1r+ZZEGxvXLDUCmXwGh0D1dqVtk5QWo9WnbicFd4ut4X6ZqOrnxXzr5e0KgjAyG/mA5O0om+jt7W1tbYVYwawmOcQDesNY79ixAwMHHYfziqSG+mPcv/yvRnRGjen0HSTVwYJrxRRBJQNZH7LbNpsNhhltbSORCOk4g8GQTqdZG1wGStZYeZYue8DS0lJLSwsZbcSeuDzEbkihEK1QD1ldXY06XT6fHx8fh9uAzh8TFCU/8TchFbA3sOFjHRB3xbYWFRVBMtvY2HC5XA6Ho7+/f3h4+OTJkw6HA+CFx7ywsDA2NibCeBsbG3jf8PmsVqsYoOrq6p07d8bjcUh4WJ/+/n4ykHSoQZ+EyGN6ejqbzYJUIF6KUJQmHhQ0E0BAWN6qmhpwLVWgaiqMnYCmZdPT00i74KhKBLpNY82fLBaLtAoaGBh46qmnGhoaPvvZz3q93vPnz1NoTg4DYoagKMwcqJA8XLlBEYdZWVlpaWn53Oc+5/P5fvzjH0O3R+2ALLzUAc3OzgaDQQhVtMVob29vampKJBIQvaurq9vb2/fv359KpSg9v3TpEmq6cl+qsQZhj0aj7KCazpbqOGyWvNJbam5cbSslRQB6foKseTUZKDxFKQ1H+0xSjk6nU7Oi1XNqJpJqczdLEmpScJo+BvJbGnPJv3PX3BFSR+S0aPkk/iJgJuELTgO1EVsIRW12aDokFLS/avmVejtCidF41uqzEDWk4uLiVCqFu3b58uXu7m7K6IhZ1fG8+u/+/fsLKr9odkJV2kaCAhX/0mybGghb7V+FUpI0rYHLSQZpeXnZ6/Wur68Hg0EiX6q94YfhnMoD5mYQFEbAAYYfvWPsdjuiTuFwOHftgPUFsYHSO9ze8vJy+lqSuBdFcAw3Yq2C5uMLA/FLX0uMNURAOrwgBEyxZW9v74ULF5D/dzqd6qa1urqKzCY6IZILwo/GGsIo9/l87B+oQLDxhMNhLIvL5SISZyjoRUkXhfLy8nA4DBTLk5LKbHUuYtFisZjoH6ktq0X0QHQExZkClRJSoM1mY0CkAxavt1adV+uwsBTI19EVkP5tZrP56NGjhIdLS0uM5/z8PBOMQirccFB+Ls/n8zU3N9Phnnuvqal59dVXjx07RqUGRfMVFRWIrpBd3Llzp8lkGhoaQog8HA7Tlh5dFLAXdkfm+fT0tNVqjcVikpSTfUJEHqLRKOIHmvyYaiA0jpuaWNOk8aWeBROM38C0lw1VxX9VEBn9SPFGNUV0kHxZ1x0dHZLQ0jDJ1Kcp81m9fs0cUx1VQWP0HrqmUbo6YUqutR+hlhjaqNFoTCaTQE9zc3MkD4QvZDAY4PuKjGVB9HzrrKBYRZUALn0pNWqOEh0yryiRVSe5ZnykeTnikaFQKJ1O79u3DwBWCvqkm+AbqnubEYa2PrbY+TWHCJDzhIhwcecxl6gSs+omJycdDsfQ0FBpaemFCxdWVlbgzLGWhGhFkiGZTFZWVtJAAGgVOy7Z0Ww2W11dbbPZKB2EoQjVbHl5eWhoqKys7NKlS6TL4vE4jxZTzk9TDC1VixggwXkEiK+qqqqvr4dJXVRUNDExcfHixZGREWY/AsGiD8m4sealuEMQXmlGrjaAT6fTmUyGvjaLi4uRSIRvYb+y2WxNTQ2I7cbGxpkzZ2iZdrUp8rU4aX3tKjVF4LyCxhp0SPXaZP7xr7SvlZ4dCGDRWBY0KRgM4lMzXMIC3kK6WnWu7XY7pEASSlNTU729vU888cR73vOe22+/3WKxHDt2TJp7CdvMYDC43W4SBvPz8/F4vLKy0mKx/Omf/ukHPvCBPXv2rKysHDly5N5773366adfeOEFiOeSfKuurka0gMgsl8sNDAwQmhQXF3u93lwul06nkUzhEWSz2VQqNTk5aTAYampqyHrBQRQbJB50LpdDZZddsGCFp3qoxlFD+5XPiL6oyCeImQBWlp/g4YKwCQlagx5oomG2YY/HI3qeenOswhSa1Kj6vwX5JJrzaMgPeji7+NoL/Lny8vJQKIQbl8lkUMWiIYO0hJY0qSgObWGRr8v9KBgDiY8rA4izLGiMShRRDzEaHCTtUNKPxWJNTU0NDQ1S+4NbjIL/z1+1+ZYO9VFRDEK2EMYuVpXrZu2tra3V19eXlpZ2d3evrKxQRkhxs+RVSfR5PB56laLd7PF4hEYChEdeWJoigvl4vV6qw+n8bbVaa2pqQqGQyWTKZDLJZNJqtdKOgMogMEdymGDQ4jTJgyTQxqIxy202W1NTkzqJZW1sMVZwdbH45HwCgUBTU1MsFrPZbMePHy8rK+vu7oZA6Xa7XS4X2pt0Is9msydPnszlckhaS3n0+sbPRG3ivhV8RhpHmNwAHijvkHiQBmYAFG63u6uri70NDt/KygreqNrKS/3FguieKpRjsVjAzb75zW++/vrrPT09n/vc55xO5//4H/9jcnKypqZmYWHh0KFD73rXu5xOJxLV5FozmczKysrU1FQmk4GBfvDgQeYehVSiEwJmRW0UMy0UCsH0IMza2NiA0ymVq8BrqVQKhVushmrCoH/IHan9A1XqpBrOqgcrX8pl5YsaT5OHCyIn+SF2MlG/I/iT9L5eeU2k+zDHOBBgjOFwuK2tTYNdbJFXLEiu0Iug6j+pjsxmn9l4E7FJpVJURYAWLi0tUaiFr8AeRk0frptgFJvNNI0EmOYD6gVI4Cu3I7xP8WkYTIABDST1Bjyy8TP7blVVFQscxc1vf/vbra2t2A1k7lFIv5qr22arac2hDmhBr1y9Gs35qT9kJWP48HzRgS0tLaXme3R01OFwDAwMUEROZsbhcMi2Bn8ZXBj5AioXKASCCobeKSU8OCwwz8jjm83m/v7+5uZmZOMpsqe9QGlpaTAYzOfztDWyWq1wqGdnZ4UjqBFegJzg9XqpxiYs1eQ0NAtVM1k52AaEIcPMoNUhosOpVIpQoKio6NChQ5cuXZqZmbHb7cePH6cYkpY6bFciQCpPZLNISDVemkOcNUn+ABPTArGsrOzChQtwJOLxOJsrDRIDgQBWW5Vs1M8QOchYglpwJbOzszTT6u/vn5iYeOaZZ770pS9VVFQcPnzY4XDU19c/8sgjlZWVd911l/SfAy4Dbvb7/WNjY1VVVfl8/tlnn11dXaUPpAZX5fNoD1DcBAWFKBVFVhqrA4N4vV6bzSYyWwRbMobSRk91sUVLWlPnVvBZyCTH4Sj4AYnGpNk8bChUcWj9RxtA7DXWRDKu8tOie4zFBzGji2Y8Hu/s7KQqWt0qNM7vZgXfQsXZbL4VvZVj45qzSdxM4n1mZgYRY6nMZsQcDgf1E2/pJ/TbxnY+L0QD1b8RX16vKX116EqKyop/ikrBmNi9ezc3mMvlvva1r/3RH/0RMQGR0xubwd69ezV3pTfBei9AdbuuG1/IV6TxHe4AQQ23hEAdpWWIOLPnEMuIrChdzKnahENC5tdqtXo8Hrvd7vf7bTYbuqPorEosb7FYAO9LS0vr6+sDgQCdDbxeL8nMaDSK2zUyMgJ+zVKR6FVKs2QQqaADWKCCRjr70TpEGqmpY8u/8CPB13ANOHAfwB8h5DAIXq83m82WlZUFg0HCovLy8unpaZSM+vv7qZGrqamB25DP52H7hUIhABB2F0mTqk+WP9HuVgPSyXNfXFx0u925XA5JAFpAUavpcDjm5+cRBshkMmDfqFPRLVP48hojpefbsn8zcfFhGSiyBRsbGyMjIydPngyFQr//+79/8OBBGnV2dHRYLJYzZ86Mjo6SVo1EIjS1icfjb3vb28auHSdOnBD9SBWWkaiTqdjZ2VlRUfGud70LNfDFxUW/33/lyhWoq4we9EoyckAcZDXwdqUdjBBR8BLAkeR+xfPS1DTDoAWLkMIxEoBynVarFVXxYDCII9bQ0IDoAglYbBYqZsJckkUnT1/0VYho19bWJLW+vLxM0zINh0wDfYgZ0ihlqy65npet96A1nMKCrnf5m/Lo5eXlbFFwqJBjY99Ci6Kgsb4ucqvupnroRp/2FKBf9n6hWgtHWWq5pY2RaG8wq6X5AJNqcnJyZWUFFWK8BMzdG6p728GsNRQC9XnrpV0lx6hp9aa2Z2bS0FnZbDYbDIZ8Pm80Gi9dumQwGEZHR6urq0dHR4uLiycnJ00m0/DwMD1HYG6IGiwEI2yuzWYbHh6m1TeGsre3F7A/kUjMzc0hrjQ8PIyLRM8OQl3cWI/HA/PaZrOtra1Fo1EkOmGkSSzGnYrSG7fW1tZG0FBRUTExMZFIJKTHrn5UBVuAryK1+3CoeWxerxfhC+YfOA+oLkaNCRqPx1FnhT2SSqWgnVEILsw86b4qHRdVrfqNjQ2MrGqsqbaQCYolcl87KOSjUnFsbKy8vJz+uchhkzG22+3YKSmw1mzhmkMNU1ToU+iePFl4OysrK2NjY5lM5v7770cizmQyYbUbGxtpFsFFXr58+dy5c+9+97u7u7ufeeYZYhdhdGmKnuGx3H333cXFxdFo9OLFixQoSUTP6pLlJ9CQWs8tA85vSdt16eSiCXTkfjkVdbaAfhLxEMwR6btcrtra2j179jz44IN33313Op1uaGhwu900/9zY2GhsbCRgTSaToGTCIdHET9wOcZKUMmEEiWwaGhpU3ro+4akRXVKxkYLMCo2xu+6/sl6K3vT5ZErgU7MSIb+yu2sYgZuZZnXWqfNzOy65+uCkokp1RPR7myYxy6E+C7b/jY2N06dPr6ys7Nixgza7ZE3fALC3Hy/oPe6t70dFhbhQIjViK7YO/Nmuri6S/i6Xy+l07t+/H5/C6XS6XK58Po/+A2tM8pNSwEZOfGVlpa6urqmpCbM1Pz+/a9eudDrN5GNYoQzDuGDDSCaTHo9nfHwceJHFCc4QDAanp6fJeVI1JxiiBvOic83CwkJjYyNJP0yeGgqp4yyNWujtIIxGnqvFYiHEwxeOx+OhUEjE8Hw+H5Z9dnbW7/fjFpFN9vl8Xq+XVKEEtpLW0HRiFE2+tbU1SY5p0v1s/ryfy+UaGhrq6+sdDsfU1NTw8PDq6irSqevr69S+DwwMULBAKIrUiVoDts2Zpu76YCx8F61Xg8Hw5JNPklJeXl5ubW09evSo2+2Gr5LL5Xp7ezc2Nurr691u95/+6Z+ur6+3t7fD64BRINIl6s+BXxcVFb344ouRSISL39jYuHDhQm1tLeU2YJGi8Stf5x5FnoyHxSzFuRZPVp8h0Ogmg8bgZIjqGTMEPTK73X7HtQMv+NZbb6UFF5MZXv+dd945MjJC0061X8xmE1L0K/L5PHLwGxsbg4ODN910k2RQ1O1ts5Poh1RjB7Y2ypu9U/Smd69ODEkCiXIcLtcWXWA2OzZrM7v1SdQppCLU8o4GZ1a7bGuABwo14OYajcbvfe97vb3/L29/GiRpdp334bVXZWVW7ltV1l7V1T3dPd0z07MRGAxMkAQImAiKACNsMSSaCphBUV5kOxwhhe0I+4MdFPXBdsiiwopQyGFJ3gQRJGQSEBCDETDTswG9793VXfuWlZVrVda+/KP71/Pg4L6Z2TUD+P8CMVGdlfUu97333HOe85znXP/v//v/HmrsYwcLidTj3JyDR3sRW3s4y8DmmtU3mgQuQR+lB/Beb9++vbW1NTMzs7+/f+PGjdXV1cnJyY6OjsXFRewyg4KcoM/ni8ViqOfgMJbL5UAgsPDkaG9vv3XrFt3bUENeXFwELmBFMS4Yx6Ojo76+PlAaSjMw3FQ8ojDlJEwkXkGlX39/P+XgAg3tCDgTkWVDtxo2IfSwGBP61UMBBqkngobNjUIFXvb4+Dimlu6LqFsEg8Hl5WWmC6lIvTKcUPXH0/2gzeQwBLx6m+yF8Xj8e9/73tTUFDoYlCY/ePCgWq1S+ba5uVmtVhOJBOC1BUmPediLItOBMwU7s7W1lX7e3d3dAwMDZ8+eHRsbo9VWX19fPp9vaWnp6+u7devW5OTktWvXNjc3b926lclk2LltuaATtmNMl5aW4HSjCQ45TIIk9KCgssl6RpwB44i1RU2FNj1OXyuCeqywkwQj6mJasiGFQqHwk+PcuXOf+9znLly48MUvfpFuNYhP0WYeq9ra2losFtF5z+VycAq1JQgSUVAldiZ5DnYIVhP1UPF4XEiFA2J4IT6HDeJYiWe+d6+d9cIRLR/XdvM6sNSRSIS0noUynnk5azG9YqLHmaiWq2MPxOPUP0/aNVbB5vE0eGK39Whk77iZpaWlDz/88Pz580TYT9Oa9XLTzsboFMg7j6TRsQxw55FE+FUPGli66sqxtbXV398fCoVOnjxZKpVefPHF9fX1dDq9v79/4sQJ4RWIqdLxlurK/v7+YrFIdRNAT39/fyAQGB8fR2qH6Uj9N3UxVKJ3dHQsLS0lk0lUrWldSCM1JJZWVlZQVmMEyDipAY1GORaLVatVukClUilkoDGdPL4DkvKDHC5wcIwp6oMrKytXr16FxQXQTC0+axt+SzgcJiNKuTPNAahXhHeIdeDP1ZRTXjOpJ27DNg1wgDkdyWRyZGRkb2/vW9/6FhWV1Citra3xsFS+jY+Pg2vzUJbS4HXHaiZ27J1AvRCuR36JnoH9/f3f+MY3vvzlL1NUVSgUYO/duXPnzJkzTJtYLPbmm2++8847v/Irv/Kf/Wf/2T9/chCcOTEEZoLP2TiVKlRhpLqZwDiC9K37j0aj5BLwRcB/0B4Rk8ROGwlQAOWB7GNGkRJjO/f7/aOjo7u7u1TYvvbaa729vf39/Zyko6ODhs6kwVX/RY+x8fHxjz76iFmkwmN8FHsb1JfxA/EWWQp0xilEsEQmb8K8ZutbPaPzWhtYvWeay5aPVwqHOtEgniMl4Zr349gxfWhTvtSmwcFV3rjBTVpQy3GiycwpXeSMHj9A/OWf3P+LL764urr6/vvvd3R0ZLPZv/W3/tbf+Tt/54033vg01D3vptr4y/UAcYd+QIED83tpaen27dudnZ2k1CgCnp+fT6fTYAWAgFtbW5DeMVuJRIJXxeoCvINJRgqIrCAdsyKRyMDAQP7JgYfC9lAsFtvb26HQgjD09fWtrq6yjEm1a0nbY3V1NR6PY7/g4bLm69Et2ajAZ+jHAfvbKqiwYnHTcKy2trbg+lDCR6VTuVzGQ3/w4IH9WzghEgoH+bKKmiq1cjSRa75xGSMsy8TEBIIqGxsb6XSaQKRQKBwcHIyMjLz55pt0Zy+VSkQb1l4f//AOHUTMUCh0/vz5P/zDP3zuuecsmxh8pqurK5PJUBk7MjLyn/6n/+mXvvSl733ve1euXFlcXNRSFwBa7+pSRxABDj9AZc1Q5XQ23qmcJgy6LHU9yNHm6PgOWQFWMm41MlUUIoyMjCB5WHMNsnmjRDY6Oor3Q/WpiHpWLs7bSpERxisik7GysgJ93mGzeVteOdC2c/IGpqDp0x5HH3N7hH4884Q1YXRZW4hhehfeUzUuonEwa6+jZr9sH0EBHwv81VdfvXPnDtSGP/7jP759+/an5FkrSVLz8o5/5IVKat43eTnwMrofoN2DWaxUKi+//PLMzAy2G/c8FovNzMwAJlar1aWlpRdffJE6BQT26MkCmwKZm6ampsHBwbfeeuvo6Igq7YGBgYcPH54+fXppaUnSS4TbiIuur69jJaE00Ynm8PAQYERPTdcuv9//3HPPFQqFhYUFLHU9dicoBNl56mLYt9SrjPqrUChEd91MJtPR3hEKh8ql8sDgAH4Q+8fi4uLq6mq1Wh0dHc3n88vLyxsbGzdu3OCEw8PDWoQqqFPAzi3JWDeYUmj4zc7OJpNJOim3tbWNjIygmg0oBAz13nvvJZPJVCqluqeaKSbvz07OreYmx/h84Qtf+IM/+AO6yAt5QLtxcXFxcnLyjTfe2N7efuuttxA+fOONNy5dujQ/P3///n3S1GIiNliWyKFQvUlnODBriqScqnEZaCw7ijTkgdU2rF7+SvfAQTUmbweNKsBDKiRRVfSGQRyEIEhMdHR0pFKpsbExYkSHh2NVVXUqwqPNzU3dwOHh4a1bt8bGxlCJ8aLtegSv1MQnMtbOsDT2G3QIn9EebFVh6x3eq+PEkPMHk1Q2+DgYt+O8Ox6t0zrH+yD2V3z5xIkT0Wj03r17c3Nz1Wr1u9/97lPM2uKS9R5JnryT+a050E6swWtTdx82LqdBBgV+cFqXl5f9fn8ul1tbW5udnc1ms7u7u7dv31YtImkQOhmGw2FoEtFoFCHNcDhMPj2Xy4Fv9vb2wmQ6ODgYHBykgwGSPTAW4NjDP4EAXqlU4FqAMJAywmsm0oSwoQVAqTflZJVK5c6dOxBda0pfQs4Da4beIJyaek4CXgBr2FSP1boT8ceG6fAxH0AdlHkoWhzs7e319vYiLgj+DvyazWYJkEUFcVILhCOIuFotASeFDTZaKBQSicQrr7ySzWbD4fDGxga19YiXArb29PQgMrWxsUGrGhu6NvCqnLyTTIBad7e1tQ0PD//dv/t3e3t7rQZIW1vbu+++u7a2duPGjcXFxV/5lV9pa2sbHBwMBoPUKG1ubgaDwYGBgZs3b1II52gG8YONACAv04UDjUOkM6RTiHOtg0GggJbOuWxXdl+0q8xmcbTlHB0dARODVuM3MP0g50CzQVpS52STEPbFtqFevYuLiwQ9eM22MNV2zCIs0CaBjxkKhSqVyrlz56TjWlMuVblxp1mXtS3WgtvPa1qeBsa62fyXOyffe5xulo7t0v0zb8ljSVLCduatt0/UIzrX24GcLzs3rIwg9SJIO5E2bz1//rxcrZrDp/SljRSc+3B+sPZLTelBANXlE+yV2cyEoDcH1IhIJELNCzWKuAB9fX3YOCDLoaEhXlKxWET3gzz+vXv3Ojs7V1dXgRRIf8G6vX//fqlUmpub+/GPfxwIBO7evYvYv7rwsTB6enro7prL5ZjczAOeDuYJpFct7Obm5lOnTilP2NzcPD09jVMmIpEVhdDP6+vruVwuHo9DYQwEAupdhLw1lQ48BQtJwBwoOUQ9hefwPQgFFD6z92jx22Q69w/XWy2rWXuOHATRAFQZyiZv375dLpdFsFVTPhjotLVUjb6XSmVXsqMJ57gFTBWWUEdHx+jo6G/+5m9aDhkZs0ql8v3vf39ubu6v/tW/OjIyws4Ek72zs7NQKDQ1NZ07d+769eskXW0pmpYZAKjalVr9PGAQpqiUTkkm04VAQHZLSwsZEcBrewmnabf4thqfvr4+Wm3IkwgEArShIEiH+DE0NORttKoKZhTVDw4O+vr6KDFDZQJ7LSPlOIkCpqGg0CAJ8Yrh4eHe3l7LobSsYea5JRHbgnv70p0UnJMVE3xc72j6WYOj6eGtvnH+pKZT75hgdaJhf2XVa3o3oL7IONTMu3iNeAP2qlB+rgtNEwrmU8qwGLjem6h5Xgcv996TMFl+Jr/H/m/jwXA4DFCLte3p6Zmenh4ZGbl3797g4OCDBw92d3enp6dRXAINCAaDeLhwktbX15977rlisbi/v9/f3w96gAAToShWOxAIDA4Onjt3jhCyt7c3kUioj20qlYpGo7du3UokEisrK7Q1WF1dhcDU1dW1uroKE1zyoY6xg/JF99hgMEhWXfNVOKZF/fBh6VYlXSf4LZ2dnawWQHyRrClmZayampqy2SwQP2HE/Py8IBo5OJDA2traFhYWGgt0eGeM1ykABZK+EjjPyspKd3c3dedwafTskHDkFDuHNwVUM7R3gvd690yt0+/+7u92dXXRBp7C8ZaWluXl5dOnTx8dHX3nO99JpVLz8/PoishI2XJwCqnYLJFXBHxA3gsbbb+PP8vLlcogmAzVKGyQ2HGShN6h1lrz+/3xeBx4hwkM34MSLepWuCXWlwBJFWoB8hweHsZisVKpxOyamJh49OiR2mmygamCQ0GwrSGwEUNXV9fdu3fPnj1LOFgvFayH4pVpzttmNNZM13Qqj8N0PqpjNz8p46jebCSPzXtnE8LlqjcPVcdo78TrRHsfraZB9z7U0NDQYyDXyxP0fvX4D2zXFTNemzCpbWeUKUH0+XzhcBicbmBgYH9/nzI8wknUl3K5XCgUKhQKzHs8a8wBqmyHh4cwcCuVSn9//6NHj4aGhihII09y//79bDbb19fHZO3r65uZmSFxxxZ68uTJ7e1tIuXZ2dn29vapqal0Ot3S0pJKpaBYoMSPqbU4rxq3q/JQwbLdeB35R7pTg0qTsFIrNY0PuAoWB9JeS0sLQkWYCVYmGSRKuTCa/BPWCl2pCAUsHFHvsACInQDb29uISkMPQFIjFApBQ8ReQxc9OjpCXUsGRSPg9IG0hzNKn2gSHh0dUdBkP0ylUh988AFedjAYrFQqo6OjcpNFb7AFMphINeKSsIZ0VmXdNFa6AVieW1tbZCwgRHd2dqIthfiU1wWTX4xgen9/v7hGOzs7pM3D4bAWEY2EYOOpFBNjTQNW/IlgMLiwsBAIBKrVKgsK5XGKX0QedZJg2qdx0mG4trW1zczMwEZVnVfjyiZrl6U9axtjWop6zeP4mHVTra95SfSNZ5GN0vA24POojbUyyTbPYf/c/uBlbT++gcOjpuafIcvVuyXHESRK+6n+VoPH0MWcT7wtme3PiqRALbxSPq2treCt6j8QDAanp6dfeeWVDz/8MJPJTE5ORqPR27dvj4+Pk5KGpYTjWSgU2tvb7969OzIyQiU03gQgRiaTCYfDzc3NaDxBLEPG4dy5cxcvXqRglxT5ysrKG2+88a//9b8+ceKESh87OzsHBgawQShBQ4LWU1j1cbpe0ZsGV4g3oQoIWWcb9zFQai4DZKbqFWxEsVhEM7qjowPnjs5nNEZh9Nrb26vVKqQ9EkSi4uGOwUf2NvRy3qYWmOAIu/BQlsFnvHXrVjKZhBmyublJZzXNovn5eQComj71MXvWNZiENWesTJ6MAsnA119/HbUQn8/3+uuvf+1rX7t169a1a9ekuuWch66G1FuJHqDGiZhspGa8B5LKnZ2dp06d6uzsRN6Alupkp/HUVCBqg2vCrEQiQcde7DJAOcQMtSUCfweds06rejzB5YAlTVUBhVTULiCmgX9gkQrnjYjBsr+/39PTg2T5mTNn1EvMPnXNaExz3muYfh76R4PjOEnFBuCDdBeQBYcmLxleNdxovJEc07GoecPeTm+6sccOPnoOapAqx0dlFHjHpAed9C4nsg6m3U+gx8F+pczMXpsVtb6+jjPb399PiH3ixImurq7h4eHt7W3aaQPhJRIJLBHcAzwUahGbmprQqBThBgmL6pMDQcuNjQ2wFFoZABegukeJealUOnnyZCwWQ1N0fn4eLzIWi4XD4eXlZV6V8D4MOg+lYhMwjfX1dW5VkrA1XyH7h36ryBGPQ/0taaDD2bAshUKhpaUFhSnHD6XdH1kmSlFo76viCzufGkMK1l7baIm8OYXs+Ko9PT0sbBFdQXthUNQkeKlJPG/HmgxLpxN0zqWJwOQXe3tHkX60t0oP4kwmMzU11dbWdu/evStXrszPzxOESanSglRyuhUBKPUiYVJdQqEDEWQqlSLXQhMTpnd3dzcFX8VikbejLqatra2PST4dHYVCAeYPZasENMxtfB2/308RGegNqZdz585hTFXbAnqey+UQh8nn83/xF38xNTV1cHBw584dCa2srq6q5bSEea11IKTjQPAgn89fuXJlbGxMHVhkBOrVdnuTgXb5Mw4Sn8CZo8bq8TmbW5qa3Zxb88cH9sq+8cYpa7ud2PViF6CtLOVzYiNYQFtbW+g3EN8oRiTS4p7l02jmaF0/3beefKWmQbBLku/bNCGD8ziNRgWUJTaoUsPJNDq9yLzOvAw3X5YWO9khLs9E5Fc4hpVKZXl5GW9xcXFxYmKiXC4nEonl5eXW1tbV1dVIJHLnzh3IEuQh/X4/sN3i4iJCJ0j6lp8cQCXEj7u7u8FgMJ1OVyoVyhcvXbqUz+dHR0evXbsGQfDhw4c+n+/Ro0cqEunr62Mpbm5uViqV7u5uShvgVDAgVruSfypsxBaIQ21TcPb1aAD1h5alyyIUkxqtEnKS3gUgU4g5kGodBw0cbEbOe0i02ktAdhYhB/gVg897J5ZSB2Ren+ym7UMkNIaQSy6YBcooDsIiQ4uUnLGTH2t8LC4u3r9///DwkJb2s7Ozi4uLVBWyRFWNLWRAoyrIC9RYtWckTuAv4T3wFJKB7OzsXFhYILWbTqfBsqHTSdgdcINBg/8DaRXXW+QK9mk2J+WKDg8PJycnR0dHYbu2tbVdunTp+vXr09PT2WwW6I+06sOHD6GmaMuBjU6aVA2UvTCxPGuK5lFimZ+f7+/vtyGaTETjykP7ZW1UEhR16sUsF8jOioOPu1PJXNbEr2si2sdES8ReE65FQfLBwcHa2hpgI+uaHJJk6PWHilfsRZ12WtZkHz/IeFyqINwKG9fd3U0kiFycldiXTbGDa8NnsUe4Cdaw1ERFXIfwhCPQ1dWFTxEMBnFMDg4O0J+kwzoacidPnoQpQfduRDg/97nPYeKr1WokEhGut7e3hxI/Ehm61fHx8Y2NDbQsIpHI+fPnYZW99NJLtC4MhUJ0bpyenuZ9cHLOA1mbemVxffTWyUjQS0HK4qorVVwjww08LSaDPHTKWACjWdXiWiH55Pg1dsJB1YK4bakO+qsGM0P1+vJnbRirl6vzwD+hOlmWGtIejTzK5TIyUnKLxCyyghgiL+Mz+v3+YDDo8/lEUuZaLBuHF9X44MsvvfQSNaXVavWNN94g8/E//8//Mwwt65fYJcejsVtglKlPAeYKhUJW10nPBX0on89PT0+n02lJ81AUx5ZvcznRaFRoIUEDFwJJ4wYQUmde2dVeKBTeeeedbDZ78+ZNqFalUgmdMkicKBMAr0H/Yrdj3qo6xhoO5gwvSF2GOdiHVlZWkskkNHObKj+muXFIPtS1O0gURkMWULTCpicf2mj1mEfjqeJFca3OLQQkekWRlQXxX1tbk64kFFub4LGwteNLeZOoXkzPu+E9fXxYmRCHe3p6mBzyO7Av6r4oLmfNUk4vFiPpDD7UbGPmQTvHs6acgRbar7/+OknClZWVQCDA5j85OTk2NobgNUUukUhkcXFxf39/dnY2k8ksPzno1vHcc8/duXMHHT5cMwr8yuUyjcSam5vPnj178eJFZt7i4uLY2Njc3BwtoAKBwOnTpynPAygHQiGxKbvDfmbrSiDeqkeXOpRbRo7+3FJKpBggiiUlf2T/MVi077Nur7NIMLIqnAP6ZIo766EmWCzPWl6Plq6ml6Uq4uvh9NGqjb8iWCkUCouLi/Su5LBi/OxD/EAKjgid8QEWV2i5vb1NEnhnZ4fdXSW8xzETe3t7NCRj5wA9s2I6ADhaqGwbCBKIX4E1JGGLDSWTsbu7i9sFOqwXpObCPT092F/cNKjo8NvY2EhncUXthfjUatkFc1EyXnqud999l2zK+vo6+DiVjYeHh9lsFp6rz+ebmZlJJpNMA/kQ8FIEJTkWTZQnBoqupC0tLVNTU5lMhj+sp35e8xAEIYsGr//EiROoZemb6j0ohU4JnB19bNcs6dCb6/Ne9zi35zWRugqwHvMBhituNe2l9vb26FIilUSny6WzHzjYfeMkrf3n4y52TU1N5NAJ2G1Brfo24f5YPKXBk2tMnboYPqddFv0dFCD39vaOjo4CPpTLZaidHR0doKKgZmS6aZIEdAubGA9a+rbxeLxYLA4ODra1tSEYsrW1hYYGaA8tZWlQfXh4mM/nU6kUVcgtLS2Tk5OnTp2anZ2NxWLJZJKc+/LyMucHmFYvLnnKCD9FIpGNjQ3OA2sCWyY+rPIBhB28UVapBspiICpIU3qWkziLRHbHJvRYxtognfZ93hyRbafJKoV4p5fooHJIggwMDHAGLkQ/Bzw4e2M8phhO3Az6LXiOEP6IKlgbuDMM8uLiohxt6DqNy8TtSmhvb8/lct3d3bdu3RocHARTDoVC+Xx+Z2cnGo2SahdtCZRWm0dXVxcVPezWKI/jxgptZ5xhVUtjjxItteuEXknpP41oAUzARii11dvk0SDhgXpLBs/quoD+kxDi5qGRlEolGFCsZdpohEIhAKv9/X1KcyORSCwWYzUJcdIrs0EVuDnmVWUTGxsbwBHe6eQwXmS/LMVI1H6KzsD32fgBPG2YfvSzG4OFoZ/ZeqmBybb3TO2C/a0ywLbUk1ZWYLDMFlK4pVIJPpvgEa/SpK1FcOh9ThWL/VCfP1UrpsBa0umYhr29Paot2IqPGX3UJBjaT8jgIVpG6qxSqaytrSGfhOmU6kU4HF5cXKTxNhRpXu3h4WEymYQLhdyHvVZ/fz8IL+ECHWnJAzDVwArK5XKlUkEdv7Ozc3x8fHNz87nnngN4bW1tnZmZAV2BD07PdcrEbRTMVOjq6kJhjkCJ12nniiNFJsE2fC5l5+0Ms0kP203ZTj6rO2FFP6zhtl5zPdhazqaDa9Wc3OR4qfkU7qnHlOlXMk39YgKBAHukVPnBdrA1EGO4w9XVVXgvtAvQFtXS0oKhfGZErORMS0vLlStXTp48ubKyArp14sQJvHt0XNm/NTIaCuJLXASb1JJFxunjt2hX6Tv0ruMN7u/vDwwM0AqdikpcYypfmJky1iqJ4n1JBRs3syYobG8eNjeBHYrqoDd8KDyBVcDqqMf/ZQ7gjEOQJVfv8/noBvtMPo+d+XZ4qRWKx+M9PT3ZbJZlxcZj4zm7DTT/7IPb3F3TL+hw9hin+k+eJVwDQj1MvNSjCOBQRPByZG14rQs1GEBr6x8ba2wH9Em1pCSKJ5rmz5g9xxyXmnegRUiyCxNDpO/3+8GLAYjxTEHuOjs7Y7EY6Rdk8IA1sQgYOL/fj7wRjgyNQpaeHHhGgAMbGxsUtiAwFolEIOggmY3YHpcmrkHDb3l5eWFhAeoI8tYUhtlKJz2gRZ+t1XPMOp+QWpGx5mZs6xAnfKnZQM+RabW/8gZAjcXPbASnU9WMc/mtcIlQKGQ73aApQXpDLifxAdWhuFTqh4nwEExw1KaAxXZ3d7E7ijPUtBd1pONjl6Ojo729vaurq4VCwe/3Ly4u+v3+V199FX49IgfCoFSTLZ0mbsb7OlTMLT47+xP7ENVevHrmJ86s9O3Uv42x0pmVQMMtwIhbbjLTw/6JkpbsTNBIUKlFcgfTrEwdzQmpmbJTwqIB0ltXaQwvenZ29sKFC3jiat7YeFKpREvWn5sPBoObm5s0tRHtx9sKq3HS8hd4eKtaLE9GnFoxhRhbGCxKxjJo2HFcN6x2TbjpEx1PPWt1+gF6U/2xwnZFBF4+Zk27XBOd4Yf19XX0UQkt2bRZLRCGAPsQ28SnrlQqAHxoGaMKwvsmYgUTJFJGgq6trY0KLlA/XHhpii4uLhIn0j8be4eXDUwM7EOdxerqKiIetCnhWmqNqMlKITj1b+oq4Oi1287iYE3MAJBQh10je83WpcXpDL6MvsPescIa1gTXZOPXS9k1mFvsLnfu3Hn55ZetsU4kEhLwFVOYBmNQFCgLxF7QapYVy3u3Oz1RHS6hKK6WrXTMA+k4pFRQHGW3aGtry2az6m2KD4s3zXuRerXl18t3YR2q6JwUNFuOumhiXitPjlAoNDIyovkGks6KswlhpdfkIenFOeCPXrfdaCnfkEQJPPGDgwOiFjWgwLgoMSDA3QnAtXsxf/b29ubm5ijulW5B45G3UZ3N0LCF01hVcs963sY1AU3Hrg751IdDeWZsraCx9kiRstWTXsw38smYbIbarrV691+PWfvYWEOfQk+dLATWDcUiC2A5D2NPbeEVyxC0OCn/7erqQqcCfIp1EolEUCGgwwBVrThi+MvxeHxtbS0Wi1EODtuJxEuxWAwEAmS9UektFovA3AjCvvrqqxBI4PyVy2V0yvHskKIG1sSzI65ZXV1lZIg5lAuyuRcekLmIKkgsFsvn83DX2AyUVHG0LKwaDhlnFpgj86h8I/sH9cG2AsVxw8U24W2q/5v8GksLsTfjqPNwNuni6+R2DnBjoMA+n29wcFBa4bpzedkU+HADlUoF3B9hWG1dli4C+g9qwWbA34qeeMxQD0hxYWHhu9/97tjY2OnTp2EN/fmf/zliqrBWZJexL3jKmDn1C7cTns0VKqeuhS6uLBRTF6PZ3t4+ODj45S9/GaWaGzdu3L9/ny/wRKANgUAA2ka1WqWXBc0cbCMrmkdD/6dLZ29vL8lMdnQ1kJ2fn9/d3aU6TLlN5Crn5+dTqdTOzs5zzz135coV0EWiXl40BlQeJREqCGSpVOrr61ORvdcDsGMC30M9UwgsuCvQLWkzaJBtO0evwWn62I8hKvIaJe+scFDZelOlJtvPgi0Of8OiJVra6unBpgsiSupLMgbgzBqxmjulQwp4ilnrJoSQqgiqJh+w5iLxPkA9vwz6CwAlDAdyWT6fr1AoxGIx5RxIMrS2tq6vr1cqlWQymcvlEGgngoPVz6qGTbWzs0PzPWgJxClU/VlWJjpNVA309PSQawqFQmtra0xHcnrspYJH2B4DgYCUGWSpyer09fX5/X6SMN7H1xDxOZgpsw1HjISSt0JMWK1wW0s8sNw4m5qwxRrShanZF6LmTNWh9eAkKoGtt7e3p6engYA3Nzd7enpoJU6coVbZgBtKVovOoX0dZ9MmsVFE4dJo+PEU9Atu3ArEjl6xWFxYWIhEImfPnu3p6eGvXn/9deQyvvnNb967d4/XoXeknDBjKHBM57Q/1zvA6NEGOTg4+NKXvvTSSy+9/PLLdNg4f/58Pp+HCNjW1jY1NbW+vr64uEhF6+7ubiQSkUgIiNDOzk48HscpZtqHw2FiX0BFAhG9LO5cUwIHFqgdpiDqYIjP2D44nEFApcR7d3d30XO/fPny6OiotuTG4wBMtL+/T06eF42CAsJSn4j5x2HpsP+f+tccTnFKvSt6xZ7YhIBKCLVt11MWrDiCqj6pt8f8NNaQH6SQsB5O7fjwXqzDYatw2B6acn4fZz86u46aHuPOUM1IkZM5DIfD+G69vb286UqlAmrp3QPhVEjyEWYVGS2V2MGw6ejo2NjYmJ+fxw/CcVMIyTRFmSgQCDC/2Rur1SoVH6q+ExUXB5y8U09PD2eud8hisgYghvMIeHbeJliaIoSHCsfsUrG9pfkmm6LTpLzePLP+dT1qprOuePDFxUV4XSsrK6TIURrKZrMrKyukpLDRgtQsSiPCnCU88SHN3hAj5D36/f6pqSmVDtZ8Fkd9rL29HTxtf38fFafJyckzZ84UCoVXXnllZGRkZmbmH//jfzw7O0t8CfrBU1hp1k/XPIGD2YtCIYA1ogi40tVq9cyZM3t7e5/73Odgp1DhRTsx5gk+Pktyc3OzWCxevnw5HA4XCgXcdqnEiMKPB40gTLVaJYlHRpQMLUVe0L0JBK1cNSGRagUwJVQGra+v37x588UXX7TvyzujLMFJCDgoOWJkigMIU+r9uSOxdPTE6+Q11RQM+P/osPnzZ+5SmsN6F0xLOV4/ne2HTzndeLFWdNA54U8b5irt60AzFg917sP+0/urenolULXoeUHyhL0B/zcWi5XLZUIw1nxb22MyOOlBwjSmjlgEsD7xx3Gd2AAJNCg26+joyOfzBK3s5+TiE4kEhlJSZIzmzs4O5dTgHmweRG3iL8ufhdlDwgGuAiZeqLqiRblsOoOt68NtZypbeUIOTsI5bXm07fdhDa586uPnZ5wrOiqXAjHVEpsdq7W1dXp6GgEWqrw2Njbw1AhrWE5smZQOhcNhi5Pi6HV3d6fTaV2C0Apdb75cKBSIgaw4rfewExjaGQK8s7OzH3744dDQUEdHx+3btzOZDIoIsVjsH//jf/z3//7ff/vtt5PJJBXYjJ71oPWmnCntyMbrc04ODLKzs/Po0aOvfOUrZGW0mzKAiUSiWq0SR0JXVQtHdT/ht3CfQqFQ/MlRrVZv3LgxOTlJbwFucmVlRck6XC78ceGq5HJJubPKIpHI3NycKjb1XLYWH1MeCASKxeLGxsbMzMxzzz2HVky9F8EotbU95gcTECeTSXYI9g9hTXDnVWNtqXINmEtNtRplfaKjnhvb4Pteu3cc261VT5LjZ0/6+H8/k176uNTeXu7xZJCZBj5mp5UT9ElR/JpD5mUFZbPZlpaWQCAwNDSEL5ZMJkUmzeVyiFoMDw9vbm7Sgo+YS+tHexGwJkKg3Cr1+2SKmIt46NTakcQnzUXVv2Ztc3MzshtYB3gL8ggwppSxYYixF4rjVDtD9p+UmkBAWygM0KQQ1XaR6OjooLqhXC5r85DagIjAAOtbW1tQKWwbB+ACtcT29hisKYTL4yvfrc/tHFVlFz+zjxLkNjc3JxKJyclJnEeUXaU7YftjiSoHjYdm9jib1BRoynV1dW1ubmazWVXA8hSktpBj18khrtjEAO+UsB3iSn9//61btwqFQjAYnJiY6O/vx7uEaPXlL3/5/ffflwoCuAcbEm8BeMqOpEUYvTMfFAjXj4GiLlEIAwwweCZQwY6Ojkql0tDQEKQs9ULU4DMzWQiUhvX19f36r//67u7unTt3Pvroo+bm5tdee405DzfO7/fT3A49L9I82Mq//Mu/ZDHWFMCTeyGwGwAHNnRXV9fU1NTJkyclsuqNp/GB1GSSbxI4koTANcR+ORlFyyhvbWl9bMA+ZmdzZmFTXm/ymUFkTSJgzS87t2QrKh1HSt/xXsUuK8ki6W9ZL1Lj8bIzdKtPB0iFvOIGefGXmq50vbFwMG7nmWnU0t3dff369Z6enmg0WiqVzp07t7Cw8Oqrrz548AAvDOpeMBjEWXOGifhRjjk+LF4zdhl9CdL9UDjF4sTgQq9eX1/H7lgxTJUmWqVvBpQZj1PPdWUsstksWVnneWvWnrKGbfChTD12GeqVYltWe29vL1j83NwcXXV0h0LV9aQ2J2bDW/0TA6GJyFxxlq6ThLDIu4KYvb29xcVFsA6KJ3EAtcG0t7cPDQ3Rxh6UlkJ/oig8bgj4Or/KnWULuChb4NzcHApwxOZsurx0Hp/Pre5HIBD46le/Oj8/H41GqQLj8VWMyp4HXRWPkvBcjoudzILX9fpqKrvC4trd3c3n8xsbG6lUSkVPzsbJipVMlVWPcoBOLVWNlc/ne/755/v7+zs7O5FpbGtrW1lZUQUTmS5mCBQsn883Pj5+7949+FfUi3mDML0LHBd23L29vUKhsLe3R09qKx/oNQUS+eJhiZIFsXo7y9haBEfR6RM5hY0PBwNwXq4XLVBYaYuG9B1t3t6Glo45fSa04k2r6s9/xkYzfb1ZyHrXeOZgOfIl+lveVkdHRzQapfKFEqlkMonoAUUo+XyemA53Fc/LDgQnoTQOAV9oduznIBjUyxB/YfQRcEB1DGA0GAwCeUNXwDmVs6mIlYtCH0Tr1o4G9R2Iw9WcGc4/RfPQVBC2hQfa1tZGi3Rq5/x+/8TExCuvvDI+Pu7z+X784x+vrq7m8/mHDx8C/7W1tVH9ZX1qDZTXpxBBSmJSSpZ6Yzr7iZobqPgFbhyQAv4sLyUcDlPzSbaNOs/l5WWYW9BDsZJoaOBL1pxm+nl3dzeXy/3Jn/zJ7OzsV7/61XQ6jZ+F1pJTekcYTkSVTCbRMKKpJs0T1IGXSztlcl7JcjsajpFyqo0YRtAzNfpaX18n2NIb0eJkTJyicHYsid/rTqxCnm4bG93b28sIoxLHZk+LH0ocKeU/ODgYHh6em5uDShAMBq08r/PGdUtEqMAaXV1dCwsLExMTVrfPbmZi+yn8UgaInUYECduWyI7J4/9+DAg0eaaEPT4dDOJ4xI6x9i5bSy2tqd7nvSvRQL1l8RbiaACn/NRYP4WuP45BpG7eONvuPY6PltDWa3t7u1AolMtldmZagy8vL6fT6aGhISQ0Y7EYHYzK5bJ6jnB7UFMhWUejUdCJSCSCNql2OR4nGo3++Mc/LpVK4+PjySdHNptF4WxjY0NdxhkpeFq2IZnQA6v8QExH13OuSLLIqoXYLKLO7ygbOCg2YCvzGB8ZqeitrS1uGH46/fQGBwf39/eHh4ep2gDhsVXjNcefvcSh7qn+G+jfmTHONOKhhIZjuGk9HIlEcKs5IexpOjFiDdUPpVAowIzM5/OkASCPOpRExTQ23dTZ2ZnL5f7sz/7s4sWLb7755uc///nh4WFA1cXFxenp6V/5lV8heIKaQqdw7hbeN5dQ82ViKSYY4yOKpHIPjtNkM0V2bjujDT10aGhofHw8Ho87vGlrg8TSlfgX/3V61NZseoADBKMJy4v/K8SZvs9somyQNJNLpVLoNXd3d+OSO7uyHlbONUAKrXYWFhb6+/ttabX3kFvNUAi3Fb3HNnXzuhTSt2j5JKQRB6aoZ5288IX9Z80PbX6oHtvCIkI8qcMh8TrmzHZ1GlL3op/xrOU2SmMPX8Bicw3cHOfmnEPPIwYFvjBNTGBP03hiYGAgGAyiko627+LiYiaTKRaLMDoD/sDu3uO1jRcG2EdVGOWXsVgMi0aRBU1YkNKGxYyCO0oLlDuCn0DDoJS2q6sLxTJbf6jHYTkRhmNWSDzyz+Hh4VQqBfOhXC7z1DWJPgrBNMhesy4aJoIb5BJ6e3sHBgaGh4eJGB48eICHiMsD3U2nlYPvUAbtf9X8AsMEGg4o5LxHqzylbUYYMSZ+dna2q6trbW0tk8msrKwgJSGZNxIhsBHI9KIfK2SWkhk7PoL47RiyeTc3NxeLxe3t7X/2z/7Z//a//W9ohoCrxGKx119/XUqq1I/xULQwxrVE2A+j3N3dTb/K1dVVsCmJsdTsn2B7liJBp5QvrvHm5iYoLRqWPHU2mx0cHGQcMFVIVIMUyWZBPOe3kmatV66pSKKzs5NMAL4X+RiGVCKXpDcHBgauX79OBnV4eLhQKKgLh6Xo8APmg5BFuQ0VoKG8duHCBfpNOyVdmlq8YoBy9gZ2XKwBlsfp9ul9xqaf/cS2EvXCGnah1eNEeP/W/nm9/KGT0qx5t+KryPTZ6NmBU2yFl92ALVX3Z6h7Yr87+ccGQUe9B7YPqV+BCUKN2tjYILoHzqNgbG9vb3h4GF4HekkjIyNoPMZisaWlpXKlLG0X/ptMJmHvdnZ29vf39/T0PHjwAHU6iZbt7e0tLS2J40kbuuHhYfjd8E+grEr9QHL4igrtK+nr6+NXRLVoAqjKjjXT0dGxuroqQ+acwY6hSkMbkCvlaFcqlWvXrmWzWbYELUI54wTRNQNGpzuXroKnYydAvT6N1s/lBwTe9Oe7u7uVSoV67tnZWbjV3LnNgMthtBuSbslZADVbfcpK4jlKk7pYLEKdpK4EVn5vby98AzwShAoI5uw5SUQDpiu2cBohWSBCKBPZaYB+1R/JxDPsCH9TvVUoFGg/pPJOhyhpH5zHVG+BmrC4FwxleJn/yOOAzgWDwaWlJWq7EItHlzgej9OSaXV1VdlviTUSVVDYAmrHO4V7RzHR9vY21tw7bdhcgT6g0opUCl7HCDxTve+olpK+Q9dxvm89reNgucdBEbypS69ltzUE9m4to1SP43UHnVSBddKfYtaIQLLXqXTYOZE9HCGSBkEEv9XmCSa+srISiUQgAxweHvb39zPp9/b2qJwsFAozMzODg4PT09PYR456oVClUiGnhJ+Lyabv187ODhAeXRZpDaN6aHSCVGjHkpb4jp4CI9je3n7ixAmc63w+v7a2Ri0WQfTe3h4q22oEpaVbL0Nt+To1YxT7V6VSqVAo3L59G3wfkAfiB3pVgDNa2I5acc1Dv5VviBk6Dnd1eXnZnqe9vb1cLl+7du2Xf/mXlaHVRqWkiBRfG08wR5TKfu5gCHp95XKZk5Nk5imol4MCqHbjkCntpYkqQLGUshcLRXeunBjzWZ/LNJN/Y8Jg3Nk/CoXCyMjI2traD3/4Q/p/Dg8PJ5NJW+xjx1yiIgwRYZ/dLRocEoNUtS0VAABrwImRSITCJb/fPzw8fOXKFZrRQOnTVVgpalEve01ymJKCvb29mZmZEydOWJfcoj1sllJSE6lUzSUsU8J5lnrxesuT4erq6qpnrBsE+vU+9yZLGq8dp8RBP0gkxAJ6TpjuTL+aNtMe3M9TzBpvQoLIjtfsfTZn19Ld1PMQEfDr6Ojo7e2lRy07M8n3eDy+s7OTTqcpWoNKhWedSCTW1tYe1yv6A/sHTyupOMhoocG9trYWj8fHx8fn5+dnZmaoOECBhJJCPHecccrBGVC8eAAQ4mtBCg69hsHJZDJAfhol6IZErywG1EXUE8TJWXmDsga0GRud0TTd9lrjnjHQJCGtm2bTIFoMDk5q36aeqEGhgd3nLRtBag8Y8Uwmg9iF2I1WbdKSeR0s0v6spFnNO1G+V1LgSNeSxkQJhC7MvFCJmUA9JmtnXRj5epTnMAhWg9ummGwZJxRPKw0hqy1eLPAdGePLly9PTU1dv36dydnf3z82NpZKpaDY635s3gmipGXpeFWS7bhJihOrGolE5ufn9/b20un02toansTAwMD09DQkGXrmRaPRbDZrkxY8CNbf8mpoLEmU7PP57ty5c/r0aa93bBEz0DwiGJwM/dYbUTU4ms3MaVwbdZyj3hWfmYGr57lbiKOm8XUQCwd4cQy6veJj3gGuGV9V6bpQoZqYtW2lqFOzQ1rn3z7w/v7+wsJCPB4fHBwE/ejq6kI1KZ1OR6NRjA4lW5wNC4tA2mPYq9uXy+XUAw12RCqVikQi+HG8udnZ2eHhYUBeWGLVavXUqVMPHz6kRhmG8sjICGA3bBPqCyqVCqRvLTyMIzYIB2dlZQVhk+3tbfo60qAdHCCRSAB5Q0ohE+B9KxpGWUkWtnSdnDFHcEN34sw2mx31pkHsi6/5Ni01Bf8aPRMnKVpvmtoTEnfPzs6OjIwEAoHV1VXp7ts2Y87qspuKFR2sqZBg2U7g8sh1lstlHgEntFgsIvrBNoyHS08mAiNq5bVbYLsjkQg6XJJ5wqtgFuGVE5mpeIrSqt3dXXZoa9z5Grq++Xx+eXn5+9//fqFQyOVytG6hFCgUCiWTyXA43Nvbm8lknn/+ecRqcN5V4sj5pR9ty0YYE3KhOAdK6EGpJiEh7ALlBop7c7lcV1fXyMgITUcB8dFr1ZyBpS4sC/YOcl3t7e3ZbDaTydy8efP8+fMKSa1nDVWJfY7gz3nXDWh/NT85+piDROFxA7+nJtBR0xA36DUuALAmjuE9j43IHUzYrp0GMLJXYvppfM83ZOz0kuwzNxg1+7OCRPEo7OCiCk3HW8TGcrlcLBZbW1tDpkOUplQqBRHC5/MNDw+Hw+GFhQVQmo6ODhrqQLzjB/QNpqenyVDfvHkT7X9azd69exchApZxe3s7sgbo0cRiscPDQ7ByCNRgQahNSbWDJF5bW9ujR4+o9VCtOTDo9vY2oG0wGMxkMnfu3PH5fJS8W9qcM5OUVMRe1ySnW3fYqe5lojfIzHyKw0mn1Az3rOC1dghbEH/79u3R0dEG5eD80KBY+ZkgCYe61VAoqM9XVlZUHOsUK9tCJLWIJB09NDSEs6nEZjwej8Viu7u7qIMODw9PT09TP6mnIOEBDmb1rEnBkS3s7OwkTULsz3Jramq6f/8+8qrk/Z577jnqA2Ox2MDAwMmTJ1FoAhQmGe5tEkT2ghnO7girms8lECjknSAMdR2KmNBQo6RQFaeOMVKGgKWKXET3k6O5uXllZWV8fBy3zzLK2EVYL9pgnmlhGh/NP0us+vmnvWNYax6OvnZNtrX3JM+EU7y/8qaX9M3Hg8hc4UtyDepdQDfHMDnenOP4yCp1dXWdPn2aCQcKgU8UCAQymUxrays8aHwiei6srq6y84+MjICWdHZ20pdra2trcXFxbW3tvffewx0mhQjPFywCZECpeXLQ+CY3btzY3d2lq3qxWKQ9nc/nQ4aClo+i+kO2FQI7NzcH8YvtRG26wM0RlT04ODhx4sSNGzdoiOdVMrEDqJkH+mYrWWq23aznRzhFup/ieOaUcvwC0cy9iNvs7Cz90qSHbp/6mdI/xz/k4TobA93osYz6kEBHShdYFg6pEhMYqWfY6uoqGzDiXOC84khov8Q4EorpnBQQrKysoCeDO49DraBBXRaR6r17925ra2sikejt7f3c5z4H4wUK0MbGBoAyJbt2RknVAPYRwRwzamdnB/DBaqKWy2VEoPx+v/r7hEKhYrEIZkgUYtEJcjzg79TxEqB0dHSgnkg1/6lTp7ydc4Hs1AXUuhpe19L2p28wP1ufAFlSwdTnNbPrznEcPNrxQW28KxttYz7rBHhNotetrvdc3s/t3T6lXnBt8QqeuW4FnzlqZN6n5dD0ooMiteAUkW9sbMCxwytpaWlJpVLt7e10TR0aGpLcc7VaffDgwdraGhqSuAnpJweJ/lAohIHe2dlBqoIaRa1nuH0QBGl2HgwGVVucz+cXFxdnZ2fhPEEZpocIoZ9WRaVSgeVNwkRyr5SJE+fSWU4kWa9nLa1wW9mvekVHQsS+FEdY3Ikln8mO96YWjumb2DDQ4l3O0xHf3Lhx48KFCxivemXuxz+83oAOCyLrADF3xPtVv8P70sZ2eHhIghr5XJSneDtgXJAZFEBgH9UjUcRECn90k/x5f38/RCPMyubmJor7gDb4AdhfwPSmpiaaiF+6dIlmF+Pj4y+88MKZM2d6e3vj8Th9dWXRNPeIAonk1OFaADRYDSt3a2uLsiCJPiKjJt4OFlkS+6rVIk8YiUS2t7dDoRA5Hsqgdnd35+fnBwcHqTdWmYK6hVksnsOq2Vj6ms0lNJ6ZByb58fMcDcy3Zd86fVvs/GdeOYui8eWOY6mdow1LZ0szJJpsA3PHgstA24IaB3GXaBlziwI/kFxYrkwR5AuYAZSxdXR0PObqPUEhi8ViT0/P6urq9vb2w4cP4U0jKEEzgYEnB7xdFNbVFprqDC6XSCRYuuvr68PDw52dnavZ1dOnT9+9e7darYLDUJfF2qPbAPGdz+cbGBiQGi/C2YS9eM08qSwsfodDWqg5+irzcd6iVVCsObb2+44Rd953TTTQMu1r/sqeRGvP4h5yK7w7BJRbZRrx42w3I0n62hYwDIIDI+ic1ouxhdeWpMGUw2DNzc2RVIjFYthroFj2WtJu6tLCROI7iL7y+iBOYEAxprLFFpvWTTI3mpqafvVXf5WCqffee4/+opxNHdP5Qcr0/C1SRzx4OBxGMml2dnZ1dfUnP/kJtfiJRCIej09MTAwODr7xxhujo6OgNGJtHh4ekrCBq0f04Pf7+U4ulyPrTgOQmZkZNLjb29t7e3sfPnyIya45V3H/kUDQzRO7YNARHIb8qheqxID3nJZTb02zLUFyPJKmn52rjVGIT2QH6wGV9arDLP3Z+YL3iZw/9y5k67A7J7Qyao8LlIXuyRlBBEu5e+VzHcK8HTXngbmACurAJXi1QHibm5u5XE5WgBQNxAwQBrovDg8PU8myuLiYz+d7e3tPnjwZjUZJRnM50oDBYJCcDK3w4HLhyJPZp68uleWAhpcuXWIRUr9A20ZpLgOwUFYOJZZe0Tyg9NTVf9pabYx+zTlxHH/BBqEW2ZCtdOCOT9GJrjHWVu+oSQ51zoCC8/b29sLCAp1hLZ2cp+vp6SHchjBX81peloggIyfs0CX0CSlZCmEIVoR+0JAT4ML6SuSH2bPxoElU6Jx6agvpOjEvqe9sNru7u7uyskKboWw2axUIiNvW19dBmZWuROGkXC4zOYeGhsrlMgVElKrCYX/uueeuXbt28eLFf/Ev/gX0pLGxsa9+9au/9Eu/1NTURAEnZhqPnoxUMBicmpqiNJ8O8fPz81ydKyaTyb6+PhoWa5x/mtoy0ARJeJXsoiuAHtaDBw9SqRRr7Znmst5vbTmYtdStP7sQfp5AzbmTeh9aAKRmoYD3aIDyec26F7u3dBErnPCUdkEajRhNndZI4KrUVfUdDi2MoyYVTAOKZz00NIQ8wu7ubjabBVmORqO9vb3BYBDaxtLSUjqdZhLn83m2ZY58Pt/S0oJPcfLkyc7OTtSHxV6YmJig/a46kOJKkKvkabe2tkClS6XS1NQUokvE7HhYvANcMDpZYKkl+2ubmUolwIppyMiiTOaM0vFzKd62L44XaevgvRPikxpub1KxATrsdZQcLJvSEmjv0CutSiL3jOAtptOGvXZJSFzMe281mX+yMs3NzWBQsteMG+RCdnSSxuQwSMTRPRZxV75sk+317IvDnyWftry8vLe3x9nwo+2fgDdKc4ZZt7+/H4/H0+n09PQ0nkd/f38ul8tms+fOnfv3/r1/74/+6I8ote3t7W1ra/v2t7+NF7+ysnLz5s27d+/+rb/1t1544YVUKgWTdWFhAcE8SoRQKchkMg8ePJDwP7lxWM+JRCKdTi8vL5Ngt52anUN9gamNIEpmcCC9ZDIZ50/qkVOdw8l8ONmvljpc7E83/51r1fuOY0/rWep6unU161FsjYyjPOx05rLyLz811iQNALkikYjjqGMdnN21plcvRVDbyuTo6AhMg10hHo+rMwANh7gumzPr3OfzxePx+fn5ZDK5ubl54sSJ5ubmmZkZurQgECzlclLYBGUTExO0ykVbFQkR4BEyPNQWIi3U3d0diUSQb6cZB2YFJ5pe1BSUE3aw0vBZWGBW4kDxIAlP77s/vhKjvi96k3LEvMuaKeNj+tfHyX3be5YCnP7cC56omQgRvbw58rFSB5Mfyv7X3d1NU0R7KvsINpvtDI73IEnIF0pPjkgkQis4NalCzIiMgloNYICgKiGHyzqxD2XTDPb2nMGMRqOtra2RSKSlpYV8XXd3d6FQcPB9wD0ShlIgQagLxxzp9v7+/vn5+e7u7v7+fgj+yCFIr5HQBIG9S5cunTlzJpVKHR0dhcNhRKM2NjZmZ2fL5fLLL788PDyczWblpfr9fqQRfD4fBMRUKgWAjrG2lsWOM340NzA7Oyt8aXNzM5lMXrlyJZlM8kk95Y0G79GqCzgTvukTGugGhrtmwtwrmOe9f2th6532mYfjsNstQfdT03N6HAQ5N3fv3r2aSVWbLLJeusp1cJSQdVYvUb5cqVTwC8rlMj4vOUNmcy6XOzo6mpycTKVSxWIxGo3m8/nW1lZ4cjhKiUQCG4rLw36uMl+aZPf19eGJC8+Bjo1QNUzqYDDIsiwWi5VKhRoZZf9plw4u5PBYBwYGUHPN5/PVahUfBM8aBSLCQ1nwZ1LuvckA73TRdxxxKFaXnWHOORvs845amD3kJ9qNWZlnnY3wQop0FsLTaXkRYNa9vb1iLouTg7HwqiM1IAM4Yps8JvT8zs7OUqkkCb2VlZV8Ph8Oh9UumakFYSMUCoEag1xzRZqvk6KQmIZTh2IXkvJ4yKkHAoF0Oh0Oh2ndGQgETp48efXq1Y2Njd7eXpo+VyoVstDRaJRmSfL6uXO/3z82NjY5OTk0NNTb21sulxOJRCAQWFpa0iRXV2t8Dq71N/7G3/jmN7/p8/n+4//4P6byG7XrlZWV7e3tkydPQnV/9OgRgQX1h83Nzel0ulqtLi0t0aMgk8ksLS35/f5isai5JO62Soq4RDKZvH79OnJ9w8PD7e3ty8vLsPqEPbKd1HQRbHjUAOSV87dvHCA2Zm/7HhvcN7hivQlm/+k4KM6XHc9XIvs1U0FWCcQ5j62g8aaj7M+Pl496Cdq/b/BIVmzI64CTncNlUFEv2Jzobqg1YnCXlpYGBwdxIrAIY2NjAHZdXV3nzp3DMSHjXFNJjk+k1IEdgeJKHSNjtLCwwHLCo3n48CGB4fLyMlNfCmdC21Uf1dbW9tJLL0FUotjswYMH6+vr/f39FO9Uq1WgduyRl6/W4Di+06HfOiqgnwL0qLcw9Akhtj505pmTirEyQ87N53K5UCiUzWbj8TiJO0nXkloUBUKHdDnsqQSJOHeO6UdeRugqbh3vi64ryWSS64JTw7AE4JIKIG8tHo/bZ3eKhpld0suNRqN0ShwcHMzn81tbW/fv349EIul0uquri8hseXmZfkPqzkUZF91SRN+kbIc2Zj/4wQ+4MfwMKINWagbQn6eAhNrZ2Tk0NLSxsbG2tkbRAFoLSBD39PTgAp84ccLn8z148IAkjV6cuiaRS0faiUoFSQaxezmhBoNGc3oiZiJgsJqawUfNpLQ96iUY7VGP+GC/0PhD+1q9zrItWDv+Iq0ppuT4Q873Gy9252w/Fb9vTFGsN+JO6CpCJRNRnggUItgUEJx5teCbFHnPz8+Pj49/8MEHZ86ceffdd8+dOzc1NXXu3Lm5uTn8a/lcDnrOJoFSjJ0BgJKRSOTevXtMaxgp6lkH3Yo8pzrCqIJAqs0S9ykUCleuXKFeDp0/INqNjQ0ejTQpEI322AYT7hdVyWKP45/Q64DrnaoK65lFB3J8agYTqMphXPgC1CDyZlIqdw5vzGtbUjkxo+0xim3Fuv35n//54OAglZ/VajUUCuGXSIafSAjiE2YrEolYV9pJaiFgAO8CNM/v91NdcvXqVawzKXTyhGREuEMAGWxiLpeDTZ9Opyny5EIUEI6MjAD3Hx0dRSIR5vba2hp+AKBwoVBg58Por6+vz8zMfP3rX/9H/+gffe5znxsaGiLHQ/lYPp+/c+cOXG8CkUAgwBfI7qo5Tmtr68mTJ3O5XCKRePToUU1n04ZTsHcWFxcHBgbK5TKtpmjFmc/nE4lETY0qO7w1kZZ6Dpk9pJ3yzKRCTbS25nesMa1pWz/FUc+hdq6rf9YkaOmTn1Ig6/E6vIyWBng8gYD6fXCAAhMwUvvLDEONQYHSxMREa2vr6dOnfT7f+fPnw+FwX18f1pYG5+Q8dVdSboNwLaqjiiHpXI6YMq4N0ppqFLC+vk7QKn0JTQJROFnesKoJ3nHYoaDQigxXZW9vLxgMvvDCC7u7u//3//1/e23xM9+9l7lsZ7NDJKqpp/ypbbRzb7JZ/MpKMjm3RzzUoBgHkcWZmZmxsTGwbAq7GdhisXicTuFiaNg5oGyP4B0xJv1+/z/9p/90fn7+v/1v/9v+/n54luFwGB4I0AcMHxGQYePFYrFisajrAjvwpNDqgVww8ez60EukWLm+vl6pVNjtSqUSzcmotMKbRocWVXdrsEBvzp49C8SPv48ZrVQq4okDBAWDQVwE1Erz+fzS0lK1Wv2//q//69VXX2V/am5uXltba25uhlK1urqKlm+lUkF6ga5pPBFgVzqdHh4enp+f904JNX23H8bj8fv37w8MDOTz+YODg/HxcTZg4gndsD1JzZ9/Jtg3OFgDU9P8ccq9sXFXFaU+qZmW94qXeYEUx/IeRymwgaX2utU1JUttENl64cIFhVeWgWspt/YHxYM2paaLSahXwCJ/S+yJa4BfEwwGEd7D06F0yufzXb58eXBw8J133unq6vrRj34UjUanpqZYA5r0WqVcmmQRRYy5XG5mZmZ1dbVUKu3s7Kyvr7P28IlgT6MUTMEkPb1IbyrdpPCHP+np6RkaGsJgDQwMnDlzhtJh0qHhcJiN5OjoCNIrOAwXskbfrsx67rZzNLD1Dj+nwd96O7Q5J1Q+jUGw+pyAWt5zirgmXooeUBse7qraZYE7NTU1jY6OEr6oJZgmGyUbVOuRRuag+wSxkdNaSXwG/FyKp7ni1NTUX/zFXwwPD2OP1F2Td4dkGGzrg4ODtbW1/+f/+X8uX75MP7DUkwPr7H9yRCIRNSRjX4fKTYiGH01yFceZiUpeJJlMkioEISTjHQ6HGRwYogQfIyMjt27dYpSam5snJycJQbq7u+fm5trb2zOZTHd39/3791XQeHh4OD4+Tu3lxYsXR0dHX3755Ww2yx+SJG9ubqbIq1gsBgIBpVW4VXxtEvWRSOT+/ftohXuXf2trazgcJqnIK5ifn0dJgtaaTI9KpdLX10dPpcYz3CJ77N/qSGBnePPPQqBevfV6/5RNd7JE9mcvN8NaaqXlHH6Xt1Gf8GsvEl1z82tww1YLWzmh1hdffFF/7KWgO3UZ3kSW96qqDVFcrJrsVCrFXEcZHdudyWRaWlrOnDnT3t4+ODiYSCRQt8lkMv39/d3d3RR9cX4a41qKq43WWT/gkpgeyouxAtVqFYuPE03TP3Sulb5QtbfYC1AGFZAiAIRYM54XzdeJItmQAoHAxMTEo0eP2EWcd+CMrTPCNQ1rTeDPDnXNc1rmUz2ihZ2dGjQ7m4XESV3Afl/OBe+FrDJuLB9a7wM0AFR0aWmp8OSAPkw9FM2DIM9RNKgSZ/tQFue1DyU+nO2ksb6+/v/+v//v/fv3R0ZGEokEyWd2IBwuJFx+8pOf/Hf/3X9348YNthOMOzIaqkzTzgF0Yzs8aQtRCy7hgTwF/Vlk5ePxeFtb2/LyMn69wLfOzs6BgYGpqalyuUzp1sLCAopL7e3tU1NT+/v7qVQqGo3mcjla0HHdgYGB1tbWUqmUzWY7OjrOnDkTiUTg1fFGqBHL5XLJZBK6ZLVaJccrmhNjjmO0uLiI2K/XWBN4ocq7v7+/vLzc3NxMvxgenGzE4eFhb28vlqsmg8J6rPxQz/PwrogWox/rfF7TAaoZszqHs9xq6pfZT5xu987njS21xcTrwdzev31qrBu41d4h8/7KccNlrO2fwHItFovkPRKJBL1a/H4/Ha9XVlZaW1vv378fCAS+853vDA0NvfXWW+l0+tq1a3A/6CkjZU4btlBhQaMQDDSLf2Njo1wuQxNEv5+tD2uL7cZdUgmvkGspSLCFUJhOIfvGxkapVIL/j2Hy+/1nz56l1pwwgoObkVnx9jCuyVRr8LMO2120gbG2/63d4+7jOWE3LWvHLRht7aOFj9X+Siw0dkcnMkBUYHp6enFxcXV1NZfLUZu69fEhv95SUyT87xyO80VHLvWM12hg5r73ve99+OGH4dBjHxCbUqlUisXilStX/sk/+Sff/OY3i8UiPYxwRXlAihiZDFbT2bZ8E01Fd47Z5Vd4D3b9A/1ptxBBkLPF4/Fyuby4uAjgXi6X19bWWDIPHz4kJO3r62O3k8h9NBo9efJkLBa7d+9ePp8fHR09deqUI5yJkxQKhba2tqrVajweLxaLKtbH1hMTd3d3T05O0m/Xbory0EHJwJGq1SpabD09PeRIMdZbW1t9fX3S7qhJT6oZ9ddbGs21jLX+6fV7vE0A7K9qmml7V84OUdPg1kyBWgelwSXsI1sEwhKu7T73OGJr+oUe9s6cHayjowM5gkKhkM1mKXPa2dkZHh7e2NigXCUQCDz33HOHh4enT5+uVCq0oMbRZn5gbXV+XQv1Lz0z5DbUUPGDqELO5/OAy8wnOXT2/q1HSSAfDAaj0SjnRyhZgWQymSRgxOdaXl5muvf09Lz44ourq6vkPL3CeI2zCt5R/YW+pU98CZsrFtfY1shYy05uwBurUk69srLCZglcoEoWr3yrnUgCo+vdM4wjUGDG3BYoUvNy69atP/p7f1StVp9//vnDw8OHDx+ivhuPx8kK9vT0BAIBwGtHktjaC04orM9WdQoIwpsGowfbAYlWtoMHx+4zPXZ3d/EtgLnRD2GNkKdFEha8xXYIaWpqqlarq6urFPHmcrlLly599rOf5QvK/VJ8D8Wb+DIWi83MzIjhqi6UQ0NDw8PDN2/erGlo6IwD/N3U1IQgz9zcXG9vr/QkeKKZmZnR0VFNkga4n3rQNOZuNtcSVLD8d0tOrSdhav3xBrwUCwYenxDi2MDGltpevWZbGTsDHy+ul156ydmUGvh3NZFWx1lTngcSBVk4IMJKpYIcWjgcjkajbW1tkP8x4ktLSx0dHfTu+/GPfxyLxW7cuNHV1UVDVRT1AoFANpvlnARuxWIxn89jiwk5wRAhZiFfiVmhgQieDoJkAGqKZ2umT2FTUX4JRam3t1fdP0U+hV+omqDHCZ/FJawSuKQ1Nw58Uc8ddhKMzoDX9KytKL71PW3raPsrXVHQmMSINWME1Vluj8PKcCaiczndPzEyNAyYM1Ct6djAlMWJtrQ2O684M907kZfx+/3xeDyZTPb395dKJdW+WhIr2l4KEXDq8ZFp0a0CdNU627yIlrdFA0DM+JBxUxZLfYiUkuFClIbxFFwFVp9ukmpPuPw7OzvJZLK9vX1tba2lpeWNN9548OABjKPz58/j8ZB3hXl9+vTp4eHh999/n7Kv4eHhiYkJAT4sAYqTSSeicQYygzKPWhwgBjA7OwuQKK0bZg7QChstzvX09HQkEolGo+gVQ9NkTwqFQjAda5pp2UG7OqxT7Pi5Bx9ngBheh0+iwyJUFnf2gr36E6vnrsIRxk1T3V7CwUkcxVpvoaO1v/b2BKPVvHkb0/xUu+vTHfVUw+1AAITlcrlAIIDiLZNAkhqY8s7OznQ6TdLp8PDw+eef7+rqos05gdXe3t7zzz/f2dl5//797u5u4YasE3rBARQeHh5ikUXkYvHbLIFq623c7TwOPnV3d/fZs2dhL6yurq6vr1+9ehU+FiKT6Kjs7OzE4/Hnn3/+vffeW11dPX/+fFtb282bN5Gmcipl5f5roGyzR2+3FO8gN8guOpLidmI53FVb0Yv1FG9HSUXvpJfJ/kRTRbkjHEkYEfv7+4ASkIK5NKy1UCiEWwpVDt0Ymrvv7e0h6pvL5VQhgoHmQbT98Ji2tNJOeNzDlZUVC6Robng3bwFEWBYr2az4wKtoiMsP9ykWi+Xzefsd1AF5xqampvn5eZ/PF4vF9vb2VlZWoNCgvlCtVgcGBmgiurS0BLEKZWoGYWlp6dy5c6Ojo/Pz82tra48ePfrMZz7DDoe6GUuAPYNKSAoawUaoKlLidHBwcGBgIJfLkfmUDo89AHx4NLylTCZD/MSXEYqwbfmcLq86jzd3rXfkfO2wFiHKK5dWEx93/qulYd+1PYPV2/N6JPWc63rbkoVt7RzgDToYMnCZPke1tIaxrpembHB4/T7dH14MLLpMJtPc3EzaAUsNPzqbzeI+sOG3t7evrKykUqmpqalIJHLixAnW+YMHD65du1YqlWhnLnyZWUjYyx4gE4ySr0BGYYu4h2rfV3MQ29ragsHg0NAQcgoQtGF39fT0lMtlOILqDbb45FhYWPD5fDMzM4VCoVQq0RhBl3DIyLpcTbpxzcOLRNf8Qj1EG2KM/b6kPhOJRKVSKZVKYBT2bVobrf2/5qA5s8Keh2fHKevo6BgaGsKpJIEWCoXUXgumENiF8mM6D14bxA+oPiqxwb47WXvdrZPnSCaT/Io/cejb9RBJFcXoEtBGmaKW3CZHgYwLHQxQBq65yJFppScGvBGYpsjv3bt3r7+///r163SRHxwcZE4yu0Dh7ty5AyUpm81OTU3dv3//+eef/2l6qrU1l8uR1D179ix6amwSkPnYSlkRfX19JM9DoZBDzbQ2DkMTDofpvbe7u/viiy9CJ4fJTlUU/BynpNv6RvX2fu/4N3uS7Z9Uw93rsDuK1d6jpq2v92XJBzmT0HEU9LPaYzmxtU02PG0L1/SLPrS8iaFgFLEJY7kqlUq5XGYpqn6MCY1aCLpOSOWCrO3s7Jw9e7alpeXy5csUoRDEKeGjgFdladwMpl8xlA1zbOBf08r4/f7XXnttdHQUoKZUKi0vL6+urobD4bGxsVwuNzc35wgTBgKBzs5O7MjExERPT8/NmzchlXvNmbdw/5na/N4ciyU2Oc4dO7O3Esz2BsPKoL42Pz9PUlQKG95h+bST4um1CKjD4TANJVZWViAvk79FtwhpGvY5Xi57pDwdMS4wVSBs7PqSRvLeubjV+jAej8/NzeE28reNO9fQawIzRIMVGyGBzyiZzOKU6F00Gl1bW8OuqUKSsBI/A6+fNDuCZZOTk5CXaVJz/fr1r3zlK7QlW1xcjMfjoMZAeR0dHTjpeP2lUukHP/jBV77yFYwp8tyoQWxubr7yyittbW2XLl2icJy1Jh1jODA+n+/VV1+9fv36xsYGVfI15wPLkM7oPp9vbW3t2rVrw8PDhKSwcYAxG7t9DfBfB+I7+ln/+pPOSYtrHec2lGLln8yTZyb/NTHsP+Woqb7PPpEjbvH488Of8fEfqzPVe1obVtSD2L14pXcP7OjoiMViKHtks1mYEh0dHcop05Z7bGyMWYUdp4hreHh4c3Pz3r172OVsNptIJFBnL5fLsPqq1SoK8XIK7ODS3ABvjnyjYBCca29XSnVf9fv9t2/fLpVKyDVUq1Wc5WAwODMzg4hENBrt7Oyk8Jd7+MIXvnD79m0YVyiQhMNhVpSa53rhC29DW+8rtDbX5/Pp3cN/sNElZxMg4LwmleCLGE5ZKf8VVGcJVd62A04i25kzckuprqb+QghvKpUC8VhaWkqlUj09PbD3sLz8VXt7uy1OAffXXalale5usrbagx02mJYoITP1h+SyQHVxzDV5ajLAhODbXBCfIxNK+xVgHBzknidHV1fX3NwcgaYqOXkR3d3d+Xy+XC5HIpFSqQT+s7Cw8IUvfOEnP/kJktN9fX35fH59fX1ychLNVTocpdNptCdtvEjqO5lM7u3t/ct/+S9Pnz4di8VQTUHRjCTkzMwM9bcAUN3d3dBgGB/aIyQSieHh4WvXrjm1Jxb/DYfDlAEnEol8Pk/CNhgMouGDv5XL5fr6+oTOOedRPsCakQYgRtPPprvh28D5qVky7vzTgcJrRqgNnO56Ztq6RGrX4NQQOZizHkGECN0kyNjWzk/rG6DS1fWsvVnXY0Iiknh22Oz0TGppaclkMiigY7JpVos2KfzrarVKWrm5uXl8fLylpQUw5N69e5Tqjo+Pv/fee+AYiO0RDCo55sBeypNYL7gmrUcVDcFg8OTJkwTgw8PD0Wi0VCqtrKzkcrn19XWcvocPH4bDYbqIsUTpcoSOJSpCgUDg7t27QLQ8I1dxqOLSUXEaEXhzuTLQTtmR9+0IGXDOQ2zO5EZXS7nWmth9zSlxzKOjoyOfz7e1tQ0ODpJxgvMQDAa3t7fHx8exI7pDSddbjI+A2jKZdDMoYxBO4Th7N2A74Ovr6319fQRAKPABIjcAc2oeNlbFUuOlSn4E/iL6kSDsajYE5KW+sfv7+8lkkhJZ/W1PT08ymYRyHgwGidWwhiglwA9BF5txYHadOnXq6tWrhULh6Ojo/v37y8vLFJqT0WHCsznhjDPU7J2bm5vRaHR+fn5zczOdTu/u7o6Ojl69ehUf2TsIeh1qDE+L9Hw+n0qlaCWDJS0UClTNONPJO+Edt9frLDZ/nITQnmH1rx3vxGusHWy6MeQrj7hmlwMnv+X9rXM/Tr/dmqCKxYe9kNFPi2Lslerd0zN3Id6cJJBUxs3uzbQm0w3czDcPDg5ocE6CG9H3rq6usbGxl19+mVYDs7Oz165dy2az0WgUtyiXy7E4WQAS5ZEtJojmxuR6K83oIJsc0oN//vnnX3311bGxMYgEdDBaW1uj10Z3d/fu7m4sFnvttdeoY6R4vamp6datW9PT00h00iQF3i51m8xpeLggPxz4/qJP2CpQHfZDp/G8g7RqHJy0teYQ3jRVgnBj1Jdaut66GbuoWO0a6pqbhHNXGKlCoYCYLf51JpM5e/bs8vLyqVOnYPuym8oRZtaK+s2L5la1u6huEK9ED1vz3nCrETBiM1ZbRcdTdmptvIwdOem8C5qCEpRIDgwLgqVmCrFXoTGN60RMNjQ0hPgfqKD6gb3++uvXr1+n0Qwq8GhP0lKDJC0VkvC4u54ckUhkdnYW7+fChQtHR0cnTpxobW3d2Nhg8jCZqbKRDjXXZUdkdqlj5DvvvEPmWaOB3YeSCKRG5gZOOlWLaEUgeM1cRerSKcLy2i+5UE7278h8/6eOzlHT4//X+XP7t9gHVr1jBAVhWZKow+LXd2zgZbl9NtoQvqpNhYNZ4dBOrNiyM+2VW5Lz9AzPWmivNzFVE/ep6YNLGoktHR+TO6AsFQCELAS0f76cy+WuXr3a19dHUwLY1oFAYGpqanl52efzYdmZmhoUvSeGAHBchHNbDFoTR/b7/YFAYGBgIJvNgthAkAJ/h+s6NjbW3d29tLRETQec7kgkkkwmcSIANxGAxYOjZkxj6GBnGGulpLyonIVHKabwTjv94MSb7KBaKlB0rU4ef6JL14xMa6bjn3mwlbJF0RqCYOj69etXrlwB0LSdYlTw5jyUXQMNKpjtH9acnJgqVBjxBAWUHdO5VuNzfbK1tVUqlejapU4XyAKzMeNH42bu7+/DZmMo8LL7+vqmp6dxmaE2Pnr0aGxsDCB+amqqv7///v374C2og9K9wefzMSb8YXt7+8TExFtvvRUKhTY3N//KX/kry8vL2Wz2zJkz4rmTyIGXkkwm19bW/H4/oguyRxIcBwlZWFiwzZGthVV5MFtFd3c33a4hCAr+pvqMLcqSkewJnX/qZwfbbTGr4+lbOHq2MJk69Xipe877tbkfxwNwvm9zXc60kf2teT/1qCl2gtlamJ961i+88IL3dp1bl5PVYBKzUVByqg6zW1tbu7u7Ngzc2dmJRqO83XA4DMrW3Ny8uLjY09NDAgSRNtIsNLN48ODB5ORkKBSiGB2y18jISHt7+9zcHNpjIAlUzUialflEWQGVY2INapOUOQNePzo6Onv2bCaTuXXrVj6ff/jw4dDQEIk4cHM8iHw+PzMzc+bMGRFOmYv7+/svvfRSqVSCHEJy5ujoCITHMYh6SXheYCk2Nez1EO3rrzmTHEvnBUnkGnjDRkExlpcmHht2xCLjFgvW+eE8qHWLejWQyrNBj4RH2MLFz+EQHZ6RwZt2lLUBc7gly5e3PEi79qScR6GKlEn0V7jDWFUHZNzY2ECGDIY4s2Vvb29xcZEUQiwWg2I4NjaWTqdffPFF1MTI+FHGPTEx8e/8O//OhQsXYrEYVhvWx3PPPUcbcuYAyq7gVJBEKTvc39+nWJza2qd5pycpYu78jTfeuHHjBq/vxIkTw8PDfr8/nU4DeuAOM5Lq1aDtv1qtAltjWPEGSqXS3NycVFlQMVRjVdX9orejPGpTU1MikaDkHYiG4bJRC5NBAa5+RRCs16Hvt5hJK61aQEt8Iykk87PVP9A3bWbPcUfscrBzyXEEdWjbEB/fxp01qdlywCXLXPOwXU2ssX6qzOsYAmejsDubd0fS2iAkVMAO+uH1YcFS+Q6WmjJCAkYWKgR7qmZaW1tv3brF3RP53r59G3Cwt7f3q1/96rvvvsvqYgaDrkggX88PeIpHKRoZxgg7S6lFPB4fHR1F2YDA+eDgoLu7m8LZZDKJJDxal3t7e8AjlOpy20tLS4uLi6FQKJ1OU8149uzZDz74gEWinjuWzIthUnxnWTvPpArVYzjZH2p2brRf5mtaftK08u4rz/Sst7e3o9FosVgkfNZa1dMh0KHve8kbmugWqrLDosyE3XSt4JRINXZjdu6cYNPLHqOXmxepJLBTbOv3+2mRcerUqa2tLVj2IyMjQ0NDJ0+eVMurf/AP/sH169cpZOcVx2KxYDA4MjJy9epVgGkS2nRLWHty0GUmHo/TPSCXy8GkojXd6uoqiq+0ImMH3djYmJube+utt2gq3draOj09/Xu/93s/+MEPTp8+zYDLOivNOD4+Xq1W4V9hWJHi2d/fX1lZKRaLX/jCF959910aI5ACJTylfIZLy4wydBiyxcVFzD3dHpaXl/v6+qwEhZ2NSoY7FDfrBBx60Lyada12Vtf73GkD7e1YbVX6GjQHsK6SAEO7TJxmXRYufmYGCDtmbcVT6NObjbV+mXOj9kA60nJO2XXVRIMpYm2HjClbH9aZDR9wHS0k6qA2NjZ8Pt/09HRnZ+fa2lo8HkcyZm9v79SpU5VKpbu7e2hoiC5cAHNI5eGaqcM6D0V9CqUfeBk8fmdnJ+mgnZ2dYDCYSqWam5vPnz9PncLU1FQ0GgX5AlRZX1/HJPHgW1tbINdUVKKaRloGDblCoQAEJNaBJPb1Ypygyenn8szp6DCjG3+53m9tCXhNEqETXdW0/gcHB9VqNRqNIsDCs7BldnZ2hsNhh1fuGOt6gULNceAV40YR5ttmfbILXlRHjBfbR0pbArMIqgnrRMw8ZK9xIXt6en7lV37l1KlTg4ODo6Oj4HgbGxvFYvHu3btra2vRaPTcuXPf/e53kVqlOqazsxO6xYULF3K53JUrV/x+//z8fEdHRyKR6O/vp+yAtdPR0VGtVldWVk6dOrW3t1epVMhwgN1Tw0nWfXd3t1KpfPDBByj/tba2zs7OFovFoaGh69evX7hwIRgMQuWG0+nz+YgLaWqstIEcl2Aw2Nvbm0wm/+7f/bt/7+/9PfBousCwuDhwyJAD3NzcpOv08vLy+Pg4mrR4Y+AqlGgy2oKMrEvhUOXsK2s1k81qe3mnh9ercH5lYRxHV4D/Oo2Sveoi0qeztY7Wc/WKzXlvw8u40yFOFMb6KUvENt3RUhGM4DxtzWdGdVcou1L54LxekJH41+Yx6CxeLpdZzxTpLi4ubm1tPf/88/fv329vbx8dHSWDNzc3l8vlOjs7b9++HQwGeW2f//znP/zwQxLZSD5RumI966OjI/gkCl4gxKjpYnd3dzabffToETH7Rx99ROTY3d0N9ooPfnR0FI/HOfmjR48oPd/a2kqlUvhNZFP5cG9vj0roe/fu0R9Hm7nDJZLJYAwduMorNa6Z+kzW5zEPG2A23qEbH7Ro2d7eTqVStjwHg0IMYe8Wg6vL2eXRwAER9MF8cxIqOux5vC3EvCC1qBE+nw9emhKwvLJoNJpOpwcGBv76X//rr7zyCq5iR0fH0tLS1NTUwsICcwY/emlp6bXXXmNMgHTz+Xw2m4WbjEzH2NjYO++8A4K8sLAAFSqZTB4cHBQKhUwmMzU1RTcifAs51+wNxA3IB4bDYRog0CF3fX39v/lv/ps//uM/fvDgAconhKc+nw/fApNEbhCMhRhINh3o+Zd/+ZdLpdL/9D/9TxQfiK7AooCpwtbOm61WqxC0dnZ2UqkU0WcoFCoWi729vV7Mwc60msax5mtt/ni6ens2erkfjonzTmznQ2tk63nW3rlqOSrOh/XiY2/tpUUy2DvZFR6/u8985jPy6dguSJfjrcBVQGIJvh2HQF5S/Jydv1J8KglNkbHks4OyqY6cCYfXGQqFCoVCT08P8F9TU9ODBw96enogySE/ls/nL1y4UCgUDg4OqFJJpVLz8/OdnZ0nT57c29tLpVJ9fX2JRIKp2dfXJ50aboxAj71B6nqkm0ZGRvr6+gRL0Wc6GAySF6XFxuzsbC6Xm5qaWl9fZ2RwunHJseb3799HDnBnZyedTqOyT6huK+P14jU/xMW2dVDOnPDKhDqKeo4pr0kEdBBwmT/8Ji+rRDfjXQD26ordgH3U20FWlXI+v99PPypcSG0StpWU8gG6W4eqwcYGDG0TFVbbxD4dNgVWGf2UFcPiSqN8ICmCXC5H/EdCb2ho6A//8A+//vWv/+qv/ipIwurqaj6fX1lZ+eEPf1ipVHK5HBqqhHFbW1uvvPLKd77zHah4TU1Na2trtAkm7wqbcGZmBmGpnZ2d3t5esnBIfABJUZ1AY0OqCtQjDfBB7j/KlCIgd3V19fb2TkxMTE9PnzhxQt4fSV2+SSUURBE2KvJM4NSs0BdffLFard67d086HpKfFjrMiwZA297eRgEcd03Nw5LJJLaCGNcyKHQt2xXIHs21UiwOucIBiB1D6f1VvfZdXnUdJ/3jnMduDzbV5FADLVnA6yU40KWKM9So83E5qZI2WgCawXYJAVkAVtjuRwpp8UcYPmcI7K2ATQOViLaBAx4Oh7u7u5mUIyMjTFPIsCsrK21tbel0mqCPuphQKFQul8+cOYMO7+nTpycnJ6XA9+KLL1JwcXR0tL6+TnswZrk4YXT4xTdhouhNUGl2cHDAQvX7/ZD88RwpP6NUGt5rf38/Cf319fXp6elcLpdKpagOmJuba21tpfuRcFJ1XHUQNPsK1fXOxkoNzPGncITt/NBJdN1PqgFiD1WB4yAg7kHHCcSsQV29Pq9iTIfu4o2OpWXqqOo4rQJ1Bho34z7brurKPinbwTsFczg8PPzSk+PVV18Vo+nixYt04FUxER6G6gxUxf7666/D5aB6O5/Pl0olul4Bl509e/btt9+GHFUqldgtYJhkMplgMLi0tET6napFaEX42lSjtLS0nD59emBgYG9v78GDB6zzYDDY1tb2ve997zOf+UypVJqamhobG1tZWQGJ8gb1PCnyZ1QJiC16eHj4B3/wBzMzMx9++KGGFBNJfan+Ke345eXldDpNDpYGe8Ajg4ODQOeWYeKFv2oebR8zViVV6DRr9lrzmqiAY74bXNRr373fcaCYmp54zSRfTazPHqLxcYnWL3zhC/atsKhwVZgZONq4Od6tw0pGEWRZmqFunZNzW8rgs5lzTwRQJMRZKnAA7ty5g3fT1dWFTinJHODpdDpN8dvc3JzP5+vt7V1YWMAlDwaD586dQw8oGo0mk0nW2NraGiuWCFrpY2T8wuFwf38/K/nRo0dIu62srDCJYcjCIgBkx23f3NxE/MTn81UqlfX19bW1NUaSRGs4HO7q6lpdXQUKaGxhBdtZvX9rox0uh/PfmiBJPcDEG4paATncW0co3XvmmgxIK0vCPykTDwaDdC4m5nCQeuaSHGrmjJph1pzQgtoFazQQnYDwEA6H+T6OP745OQyaQ+rpWlsft1L6m3/zb/77//6/PzIyQuwVjUbfffdd9mPmPFYYhqKiTIkr9Pf3/9t/+29Vyp/P58HouC76M1gc5QwHBwex+Ogo6T4Rb8AtpXqAMOg3fuM3/vbf/tu/93u/9+abb3JR8nsHBwdLS0ubm5tf+9rXJicn+/v7OaHFiPGFiY9bW1vJnxMJ6dXzaOfOnXv33XdhsqqkXrGg9ULa29vv37+/v78/Pj4O/MqbBRhpa23b2X3K7fF6x/bwhnTNTw7kRzizl4tdMyK0R80v1HPM7Z1414uTEREC/IkO75PqB1Dip6j9m2++KZUJTXQbU+BNI+2PW80LFqSgJJ6suUNn4aVq4VmHWikFpHv9fn+xWKTMAThvamoqFAqlUqne3l4UbVj8qF3TWAAlNspt8HB9Pt/IyEg0Gt3c3FxdXaXCDT0p1HiZZLKnXV1dkLt5kFwut7KyIpIT0TEoB1E8uqmEsRTUUtlIXdzGxgZkbaBAXHKIJXAh9CZsSOUQyBguh/j8TKDD2q9PZ6wJL/B9nPjOm7GxNtFOWYtTA5d1dnaOjo5mMpmBgYHe3t6enp5sNovdEWCNRKccKBHJHRfbu73pYW17Ru89y6wT7rDm+aaivba2tomJicXFxe7u7l/91V/9xje+8Tf/5t8Mh8OVSuXq1av37t3b2Nj4P//P/zOXy7W0tIyNjSWTyVwu19vbi+iSTcwy5zc3N3/t137t29/+Nk/h9/uXl5cBbeiESz36xMQEGnUY5VQqlUwm0XrM5/PRaLRQKMA7pMSGxB0I8q/92q99/etfJ++9sbFBz9Jbt26BzgWDwUuXLvX19Z06dQrMUGRnyekwAgwIsh52AvBmaYl74cKF733ve2pArq7zamLAQW55c3OzXC6fP3+eJc9GSNtSODCa9jVBDK/p5OA+Qc+0mpw1ZZlzdjJ49wDv+e1R00w74KHD725grL3FPvUwRuX/rL7m4yQD/WQBiNGIsMU5liBlb1cvUlWCupI8a3lVyjGSCQEDwc5SQlIsFnFaCVEDgUClUnnrrbfW19cjkQhvnVtVumZhYQEJYDhzbCqpVOrg4GB+fv4nP/kJ/Tjk4xM2qlYQT4qE++HhYS6Xi8ViDAJKkswJimJY2/jLqnfim0tLS4T5JHzIgN++fRsL5fP5qBLe3d1dWFiwdHckn7TCbUbRgmW2q1YDM+2dTDVLZrzf94oCwweviXFba+j1W/VXYEpQsra3tzE6kBxmZmZaW1vHxsai0ejS0pLXX4ZCw/NyHr2OenGlbbJnwX3mkmWeAFMgn61Gbpg8kPSzT45QKPTbv/3b6XSazSORSPzwhz8sFApzc3MPHz7863/9r+dyuUePHv21v/bXLl261N3d/Zu/+Zv/8B/+Q0hNajNEOEhc+MILL9y4cYPcI5JVTJIXXngBdd90Ov3cc88RSqbT6bW1NcrxJyYm6Mby8OHD5eVlBBva2tpmZmagHv3e7/3el770JbwoxMEPDg5ee+21v/iLv8jn82y9PT09/+Af/INvfvObs7OzmUxGbSdJn7IEiGNEeMXNx/nFMqLldPr06T/4gz/4oz/6I7/fr+691sQw7MFgkC56c3NzP/rRj2hbSgHzo0ePaGcTDAZfe+01wmsiEkJqBzGgtnPvY3FNp57LeoTeeV6TFlLPNHvnfE2qia5rZS0s1a9mI12HdFTzPp2DuEHlyo896y984QusCsqiUOu36W/nFq1enfNPJ6Vjo2aKCBRYkcFIJBJYK5rIQCxVXpuA4vDwcHl5WUWPClepD+Qg9awkFUp7xLwaKR5QeQyn+KK1tTX05Oju7gY37+7uTiQSnZ2dNBKEP64KOs1jFW4oMbK1tQUTQPVg+BRtbW2lUomdibvSO3BSFsKIbOzvNCmu6WU7dryega6XbNRA8fYtNGm9ADt3a5pOmsyGQqFoNEqpyKuvvjo8PIxOE6wbsLVcLodd1lyyj+BM3JpcRt2Sw2Cxq8IuHjFKfT4frhm7L3L7p06dGhsbgzHd39/f29t79erV73//+z/4wQ9mZmZyuVxzc/PXvva106dP/+mf/umv//qv7+7u/qt/9a+SyeS5c+dWVlYgdJJghKTU2dm5ubm5sbFx6tSpd955JxaLsUuBTmSz2evXryNYRoverq6udDrN+C8tLeF+bmxs4CZPT0+TiqSxelNT01/5K3/l3/3Kv0u0x9yrVquUJpw4ceJHP/oRNSx0Db1169ZXvvKVTCbjiDvavBRDx2Rm0ckISlJmYWHhRz/6kapRbBSl1yfCWWtr6+rqakdHRyaTobkwRQmUCPX29lIZpMan1o3VzwCzOx/XWHLzFOWq75p99TVJIDWPZ/7Wux/UBEwsFGx7iNtTNSiDqDftCS5ZLI/nKogbaJrEARw32dtNxwFW6t2Q7kCS81g9AqWFhQU7vuijPv/8801NTXSzhvqWSCSw0TLfykGpboKaFPZnclZwRXEfkG5YX1+30vJWY7q5uRkzyjwjf0i6n/oCOmezFB05OjU3ICnEnMNZs/lDpi/9TGEKWjfB9vRzaEkiiXttVr0EY+Na0+McNV2A4x9sq11dXclkcmxsbGFhYWVlhWJ9DvZa9Zdylo1dfsxPXAyFese8DRsA8U+gWDZpstDMh1wuV6lU/vbf/tt9fX3j4+MffPDBd7/73VKpBFXua1/72nvvvTczM0OOtFAo0AF9YWEhGAx+7nOfo+Ug9bpCujKZDFOuUCi8+uqrh4eHxWIR5gbYNO3GY7HYo0ePKDiMx+OJRAKdqYODA/IfVMM/ePAAnxefNBqN/uZv/uabb765sLBATa/wxqampkKh8Ou//uvf+ta3SDayebz11lu///u/Pzo6CsFZVkBWhpksVivvhQkARkxQixIWrR7qjTwkV+IqiC6sXyQHaPyGHHl/f78waxkfb875yNA/vLi2XFdNP4djd+yZ+zPX0j/tAnTI1A7+5mwSDZZPTUfHS5mVpXpsmggz8afwBerRDGtuF85tOakwAToHBwflcrlUKlUqFVUbQqjIZDJ0rxgZGQFzUPUXLhgMP16npdGIRqKHAeXgWpubm4FAgEZKzFeR0hydUmjd5AaZN8xLVCzg/EHo3tjYwO7zh1JNI0wTfESeSoUDmlUQE0VwtCkaW4buEPKcwgH7Uhv895hVUt4Jah38T2f0BXmxnufn55ubm4eHh9F4I56gTxvlyFYKVY6C5TXKmVJVoTclqx+8j2zHTUVJeI68Tdy0vb29f/Nv/s1/8V/8F9///vdv3boFIry1tfXX/tpfe+mll65du9bV1UXly9LS0osvvkgn5a9//evxeBy0Tf3LiR2np6fJOtL5s7e3d3FxMZfLsfETv66vr2ezWQpe1NbrxIkTOzs7X//610ul0sWLF4ne6OcSiUTW1tYikcgbb7zxpS99qaWlhbaKg4ODuCPwnaha+I/+o//o937v97R+T548mclkpGau6O3xGj/46Rq3BoE3Av0JT4gW9Z2dnbzcevOE7qmpVGpraysQCPT09Kyvr1MmRqFDKBTiFahLqlOdp0moDvFtH4s6WEvtaLfaKSGL5P2CV7XDgZ4dNpG3NYHjZTvMwl/UoYDjKZOVsVCRoQVfahpox4NroH2jtqq7u7v3798vFAqdnZ19fX2pVArXgFoYvhCPx7G2ElAHlMAiS2OMHQVUQYAamDtEpc3NTWpPID4jL4lLjjMLatHa2gpIQuEitrWnp4fVhcozrGpQQhp2oIYK5ctuYzwCoSibEIUetAXp6ura2toiY4lLxaYCAsCjra+vqw5eMqq211RNs2stl/WpG/vF9XZy66qEw2E2Gytk6Pg7ji9jz8+MAmgGAWN7BnE6efIkTPbTp08vLi7iwdUrWpPH11g5XjUjjoSe84MUB3HzcQKwAqFQ6Dvf+c5//p//5++//z6fbG9vt7W10Svjq1/96l/+5V/SWXxgYOC1116rVCpbW1sLCwtXrlyZn59fWlqC2Ynz2NnZiVFjSu/s7Pz2b//2//A//A/44Lo0D3L79u3p6en9/f1XX31VMDqEmWq1evXqVcz0c889l0gkpqen+/v733zzTRBqWByIVpMv2d7ezufzc3Nz//Jf/ktUPghM+/r6RN5Q2ffT99jS3Hz4eI2Ieye9GqkjILXGQA0PDz969KharZIDUEGGUEo2bAoy1JMapuPW1haiQAcHB/l8npe7u7tLCkcgoWXd8efNTwy3Gp/qk5ozpyYwIkfVMvobuMM2d+otctGf18QGaxK9dNF63/HuN/Ymf8rGq7lB1TPEjb02wYhAAVAyRkZGhoeHQSdUCmEhG4ymiH1q/6zT8lKtByreNzru6J3D/cCJQ6xA0jCSVKWSGC+YaBSSbDqdTiQSsLYLhUIoFOrr6yPDw3Ulzw8vlVpz6r40IHiO4pLzyFLvY5U6/b00R+2LsID1Jw3lvG+k3qu0MZ2mkeCmT31dNXJcX19PJpMPHz48efIkihNTU1OI/tgmsw5eIVaZN56wt2Q12KzRr5cmEugk/RkK4mlR7/f7f/SjH0WjUZFS2traLl68GAqFzp8//x/+h/8hZ9jd3V1ZWfnBD37w/vvvoydTLBahDO3v76fT6c3NzYGBga2tLepjj46O3n333d/5nd/BWNvGmErCb29vv//++/F4/NSpUzw43ZmXlpbu379/+vRp+rmgk97d3X3q1KnDw8OpqSm/39/b2zs3N+f3+9fW1i5duvTDH/7w/v379PSgcTu2Cae7pkiAFD+IM6xonNBqljPr4nd/93f/q//qv1JJpDx0Ce3CsALnxNbbGcjX6I6tsgx2BX1HxtHJaR+ZlrJWMdXBPbywtf2CeGiOzle9xOMzTWJNnPrnPxRb88/HHmXN9OUzz+KFPhz/jhfJC56YmAgGg9A8GGs8mq2tLcvvljvJa5NSO4eNNawQDAYXCR68IQGUlLNzZr4M0RuBNOgBtCilWBG9YxSTuT3mUyAQ4DxKYOI0CVcBCZE/KLFjaVhjryuVijXWuiWbjXQGWWvm57fXDX7l/Fbqsj9PUQw72dLSUqlUeumll3K53MbGxtDQEMki6OdjY2PqP/DMw4vX2+DDfu49+AJScxb344VS6XpwcPDnf/7nv/mbv0kjTVQyenp6/vW//td0otjZ2YHMcPPmTRCVzc1N2g8yddvb2zc3N4eHh2n7jbNPne3169dffPHFGzdu2Eox7ZHNzc3Ly8uXLl1CnIDbW1paevjwIVqPsViMqsJEInFwcPDBBx/81m/9Vnd398WLF1E6vXTp0sWLF3O5HM+Cs2IFgRcWFiSJ54wqToyExiT7znMh68HQEYwODQ3x5WAwaLllapLX3d1NXtGRwCU+1vqi9RJDvb+/7/P56kVOTcaeWvktxy57v+zlLFlD7xiuetd1cp7HMc3Hway9F63pMetXTzWUm37RhzJmIHFU+mHRKC3Bg8b88U+nC5nUTa0EHUElUAlBE0soHo8vLS3B/7M9mQgSKfCBjQB1DyYpkVo6ne7t7UXLgosSzMpN2N7eTiaT+XxeYkzcgAqRqd9llsv50rNIuRFHW2xCO1zsaqwNi/t/0n20ngtwzKy3fXfybZt+joMtKhwO01YC+nkoFCqVSsVisbu7e2Bg4PLly9anFhzkVZCw4Yisgx5W0LZj1m1wplJ4ABnCOyJrKlFnZmaQiIOahlRpJBJ5//33Z2dnb9y4QSXUwcHBCy+8cPr06WAw+M//+T9XLxvqxanYovstsyKTyfT09Lz22mtXr161Q21RrM7OzocPH96+fRsmMrnEpaUlApRXXnnl8uXLagd87dq12dnZvr6+YrH4l3/5l9lsFliGdqaEhlTzi9w1Pz+/urrKiNmZplGSxoDT4RNl1EAgsLW1hdgklTvVatVWb8k9Z5ITjzIIuopqeSQEura2NjIyIn9cXradtxYgbvlYTktZsZpT1NtXV+cEopTTWi+CrKna0SBveRwxdLsVOeBJTdjEGYGnUh52N2t8Se9v6yHXWlSUoYOCgUIyUugz8E81K+EAwWBeFovFWCwGQAYzDGuIiaStHDx/5hBWQJYUGIRGyyweUJH+/n54JtFoFOEnmKRI/XGqUCiEejqaD3A80LGsVquJRAK7Tx2zkHTiX54rGAyqexYUCLA5JrTeh8NLU/RQs53NMY8G6V+vv8DIgPX/tLz1SQZYUoVOU3PntPqVfRYGnywTQs/0FCfNQANiG7equ7HNg+Op2XPauWefF2/X6XeHaq4SQTJAFFWT7KJrDNOpq6sLDIGXCJvo7bffpukJmYm9vb3+/v7R0VHcBRpCdnZ2Dg0NEXhRd4NVyuVyFy5cWFpaunr1qkN9sQPIBJudnT116pTYTXi1a2trSCnQMBcPdGlp6caNG9lsFuVSWRAFo8J2WXRdXV3Xr1//7d/+bUHDWrMU48jwCUpi0jY1Na2urqJuRvUv/gdEclX8c/8SqlavA3uw6nHX9vb2uru7Jycnh4eHicKtIk09RKL54xETgaSmr+lYfJt41xzzNhRtkOOxmDXxhxeRa4CWeFFv55uOk4FNwMrRavwxGQEXtekTHsexILb6Tt0TVCopcqU3pqCSTSKWSHAADqKzoz49PT09AwMD6BgkEgmSWgS2iP3T9AhoOxgMigcSDAaBp9vb2mPxx5AIFl8ONd+kxJyBCwQCZAtJTHN7mDAVDtC8plAoUJhH4pE0KavRtlevmXz4FC/imB50vV9pt5eSDiunZvLwUxzYu93dXdppNjU1/eQnPwF3mp+fLxaLkpwmAymI055EVturaKrvNO5NbgeBKUfLeQRFAc2Ylh999NFv/dZv3b9/XzGvpQ8zUOQPd3d32bYPDw+pEiBRhpXESezr63vxxRcvX778ne985/Lly4SSAr7s44RCoXA4vLi4+OjRI2ZjLpdjuZbL5bGxsVgstrq6yh8WCoV79+6p86ejs1HzaG9vv337Nj/Yz3d3dwuFgjLwdptX0TlmmnY/NO2lZh00g10wEAisra2pEtgZcw48KuY/RFiFvNpieeSaf370cQh4fIzYOrN2huhUXpv+TD+35uFktr0wiPehHCjPK/7Hh+TDHhtr6XLUc42PwwPR2NVED/XW+SdYntRCvBe1sl4s1FgshmgkicFisZjL5YLBIAnr1dXVBw8elMtlqlFQuqlUKriEkuXUi+G6BGjQMADTMbXaWtBkoAECBTKUXBIvs7xt38LOzs719XU5IwQNkkXFbxWuZ8OIesPuBEqf9HDiuJrvxf5gVQRsqueZx9OeQ6auzPssFDeVy2WK/mnMGg6HfT5fNpsVwVxJKmcqW87vpxgNzSXbOwaUAD8UFAt78W/+zb/54he/SJAB3oo/KGlfEIY//dM/7e7ujsfj4LZoD2gkq9XqyMjI4ODg97///X/4D/8hxSxqH27Js4rHq9Xq3NxcX1/fe++991u/9VtHR0coJYRCIf5kZGRkZWWFysCHDx8SA4G9qE9gg0E4ODh48OABDcPs5wpqrQCW5LGYrpjU5eXlVCqFzAuejZSGIKS3trbOz88rIVSznI+BIm4mXbm8vDw0NFTTBHkdmqZjHDUdcxHs7K8cmYSa0Mcz78p+00EU61URH/MpVPbxFAaxom7PPK+Nne3+8Mz1o+pkkCklFb1Pywum5zR+aKVSeeedd/AmlMrDEFPOA7woLXCVWgFKkAnZ29uT46C8H8MKis1Ckuby/v4+rFVES65du8bSZUYSBvK30ADoZhAOhykEIP0I91Y9GfDdrHCKHdhnvrnGX2j8V173RAOuzUkVvVZdwKHufYp70IXwNGnoNzk5eeLEib29vUKhYENUb6xg3SjvfGvAVqp3JzYFwrSBuKLKJmRkhoeHc7kcCWqcR2y0pPIw8c3NzX19fWzSpGeCwSAtje7du/df/9f/dT6f7+zsVEN3TFVN8hXsQ2wuvWDoD3D58uWWlpbvfve7yE8inaqybJaJyvHrHQSjHR0dly9fdoz12toa5p5WvHbkpdpBk2vapYJEg8+wvUlMGEErq/ciDQmdlmYFdO8jZ/Dw4UOkz4ViOYhHvXd69Kw40kkz2pnTAIluAIbU0z61k9MCFV4P2utr1+vlKGOtSOixsXPaOHHYNJ3+GG0jTXdeJAy2emQpJQFAkKHly43CnG1sbOB4ViqVQqHQ3t5eKBSe3p9RiYJdj5sjW0/kHggEABwwr7jPlEghgo6MulqXkiOiD+TTTOuT7J/SmCwMlFLUQhvsFUiOfP3m5qZqOqzWEqJUkAK9r43nAg3HpZLeiDWR+kNla+0nVmlXnysf4Iiyi26oPusqF2b/2N3dBbelQkQG1GJZDjho6wMt7lwvBFbzqlgshsP48OHDfD6vwiJvHGA9Ym+I4PxcL09O6C2xQw0jn4BmQPGUIvN777335ptv7u7urq+v9zw5Njc3k8mkxANoWw5rGL3ySCRCzfrDhw+//e1vT05Oqn0d0tXEXnIdqDDgtoljUIiEOoKWJCqjtOz6kz/5E0urUB8/vUrvg4ufPjg4+OjRI5bbN7/5za985Svo2FDzDQrBzVjFQe1neDmiz1MBXyqVEAAREwaKqgRYaloDFmC5XJbr5vP5ZmZmLly4ANfFFlV63yOHstDgZhY6ECvcyY07/YO8iLP3ZwtkKxIS5mN1vS3wbbvGiJNG7G6DV1shoTIlZw4T0IBDPM2Nq2eobrReXGAP7lL8yprPbEc5EongOHR2duLGUmdIKUGhUOAuldXp6+uDwiG/ho0BCh1eMwk6anwBQJCUZGGwqfCoyPJFIhFQY3EKpRkis6XKIHjf4guiQcPoE8FxObYBMFkJlZFA29jYWFlZAaomZ8XN8FyENgKgrPaQHUMvQ8a+FysppclhWz3Yz/XKyAdonjG2LELMriP4ZzUnf55CdnFpEahjk4azpQZDDTLVThyg56p3rXq3obXHOmfEcBUxBIlE4uLFi7/7u7976dIl9HVXVlYUKmmfpqD8hRdeGBgYKJVK165d+9a3vvX+++9DhJA+htaO8GWxNYJPjq985Ss//vGPr169aqFk1Dz8fv/q6qq63zHH7LNIkaNmr0Lto9A2SB7cvXv37bff/upXvyryBjU7bNK6Ye3WxEPSbSepKCFysbMoblSVqWBP8joyXnrX6lPDgdfPzdg3aKfxzpPzC4AlT+gsCpshs/6yaOCy2g2sllcFW0xcDQ7+Tc0zOOuXrW5ra0vQs0MosOeh/ZueBWv50z4PXqzZcd29u5yupD2w8U5ImLm3t0c1YLVapXE4L7KzszORSFDpzs3xOdOC2BPTKZIfU4TeccDNuE7d3d1oMIJ+sDVJ2l/4hm41EAhQYiAXNZfLqYSdRaX2qZQjgjtLBIo8J6kStg1lEfmE+b27u0v0yqBhrKEcECt4fUYNtXVybUddecp2wMXXdoI72ygId15mWhRJGWunG8szDbTVM/HOAf2slqxHR0cvv/zyrVu3qDhHmJ+RaQDlNzi81/XS1blJFStLIoa9XyKIBwcH5DynpqZ2d3f7+voikQjS20w/QLOtra2TJ0+Ojo5GIpHr16//s3/2zy5evMhkg0JDPxfFOlTMIpHa1dX13HPPDQ8PIxg7MDBw/vz5b33rWxcvXtQ95/P5RCIRDAY//PBDKmD1jjTmDXxqZxygcCAhubGx8eDBAzQmmbrSDbf4MobYdrMDICKkAA5Sx1T1h9rZ2cG5ttwGBV7cA9EAOQBdbmZm5qWXXlJIp3uw/zz4GDxkowVOcUrD5U9YdqwqGOR21JswDnjijRR1ISePYv8rpo3lODmRgb0BWx7oQCX8QNz5VOfd6VDnZGMb7BtsOBT1OUPsPCHUusPDw5s3by4vL7e1tZGZIfSz1oF/UpxC6SqlZVSfE3YRK+3v7w8PD7e1tS0sLGhh4AUIu2BWcXW6U1MUiyI2qLTaSmE0ZeixthTa8IxKCuGSHBwcUDQP1kFSnuvCyGYlIDaG4Qal2XxyVCoV5OQTiYTf76/pDNphlKGxdJoGdsqrX4pfQ+Rh04naxmpyeJp+QUdPT8/S0hJUSxAGGp1YgUZ9WQ7IJ3XnbUKSwzqM9psEE3JbcA7Yqnt6et5///1f+7Vfm5ycXF1d7ezs5MuoBvr9/pdffrlSqfzgBz+4devWtWvXKpVKOp0G8w2FQmtra05oXKlUgsHgCy+88Prrr4+MjLS3t9+9exeyx+Tk5NmzZ7/4xS9evHiRBXl4eFgoFPL5PDAF2zk3BgXWSuXxz3r0NWkhgDKxtN97770LFy68/vrrTAnBIDWrk/jzYDC4vLwcCATUXUFJcjuqaDywPG1/NYsPPCU2/Gy+anFx8YUXXrDG2juxW588Lw4Kq6+eZ+24j/LPVOtgFUW09r1Ys2AK7e52a+QRZIJ1z7YdKK4eSjjYIkkDPbOAywrcPx00bTX6S4uh2NtiTohBLEEcSQEQJzqsRosTTU9P7+zsjI2NobgGyKX4V0grqUKZNpSvC4UC74kn7+/vn5ubu3nzphYSsPXu7m40GuW06kfOLEeIQEl/HD1HKHl7extsHfV0q/sBrwOLLAYIaDs/A5twiMArCioNSfHNMeg84+rqai6Xi0QiiUSCInW9DosMqO+UfZ3exLdD4bLkJH2tUCio0bD8FAcbsf+1c7rm3Dqmzt/c3Bxl/V1dXRcvXhwcHDw6OgLfVCMxLZh6J3QAEGfJ1SNO+Xw+dCWtYBZzAwarzCsRXrlcvnTp0u/8zu9sbGzkcjm0Et94441gMLi+vv7BBx/8i3/xL+7cuTMxMUEvCxpZYQjy+TwwJZ0o8MR/+Zd/+ZVXXsGzXltb++ijj+DA0A50cnLyK1/5ikSAafShDqW45PSyQPpZZCpvOzTnn9hZbDR5msPDww8//JCWI5/97Gf51fr6OgvZtgKQrJtmCGu/u7t7amoK/iLTxpLMbG6Qm6SqU/41P0MkZ7ajj8YJVYGsCaz/tn58NrwNuFsSOXnKl3gSAThFZ6qVV5W5XT5S67RDRzERXiOEYBYXL8gxtd7cnsZE8D2aKnYa66jHu8PAWqnqxwGd1zo7g9WYEGN/a31qLXhgEGxfX19fMpkEuYPeL3hBxtr+19KbJOoYDAYTiUSlUslms9S/oq4HKnf//n2MNbESVFCS3dwYhq9QKMDuAEzkEVKp1OrqqqI8fE/wOEHqDtYm2MRC/4g9aaeF1r63t4cgBptKqVRaf3Jg9GEoxuNxL7VItTze16kwsGb041ULszXB1lKTdbR/+Kl5cvWOVCqFQ53L5aLRKOsK6X17V7oBvfpP4d1bC8Ij0DbFORW5PnURI/AHZITEmclkkskkGYh/9a/+1d27d6emprhhv9+/ubkJ619vhynH9Ovo6Hj++efffPNNv9//6NGja9euyVRlMhlyNqQo0UElX4ez0tnZWSgUotEo3S1eeOEF2lnUq2i1sINd+X6/HxOzs7OD5omC9z/90z8dGBhg7YgvK9fNMV6k4rHXOzs77DQ2bVvvLTikFxAVHDtEplQds7y8PDw8rMakjaffwcfcJMGkEqizv9JJiEi8jCabqAce1HhSNoHyIves1t5eGqiFROqR9uz01qKu2SyUn/Et+MJTATuR3uoNtxeSrolOCjOV1batpin0og05wgWMCEgFCTrrWbPlKtMltQ2EK5PJZCAQeOGFF5Dr5Tz49eLkwd8ga0QcqjBKGgUKBWSsi8ViuVxOp9OHh4eJRCKbzUK3YueXEy3PWlCy49ZZxIp3DCQi0AlDQH6SOhqagTnKNXxTairO6/cm3J5JlrdRoVKsaoLsXRuNUQiFk8fxrNnAoA9RW4QBxVURSqPvHx/9sLwUb6qNecXbl8ouuxHTjyAPaVwMEFv1Bx98cHR0dPny5du3bzMNgLwI+PA5wuGwKmPhjWhlvvbaaz6f7/z58/fv319dXS2VSpR3vfHGGxTllkolFuHe3t7c3Fw8Hmc7wZlg0rKIFhYWBgYGFhYWvBytxow9qK6Hh4exWEypqubm5n/0j/7R888//81vfvMb3/iGuvEpLPYmqEHP2MP29vb+7b/9t4QRNgutDHzNAm6tBdxYx32mu9Pw8DDloI6V986E/Y8Tle3t7TT0IWPEJ/L2nPie0MFGwLp5J5mvpgfd3d3aP5i0hKQqi/caYu9TO8lP+9Tk0hyvVFGRM5Of9jSqx2esubN5k6d6PV7Ek8f2+/2o2aGp5LRcI2HoQO8yT0ot4nFEIhFGjWCzvb29Uqn4/X4y5uzSmEX2oa6uLoJZawi6uroghEYikampKT0Unb18Pt/CwkJfX18sFgsGg/39/fPz8zdv3iRYVhsdXQVI0QHvNJISiHC65WKneLpAILCysrK3tweQrYSnVpG3NZH++UkZGkrcMyFwq61mppeF8onOX++iDAWbK+sqGAyKTuDV2PtFHYrwJLjK5ySNVWBNwgqPj1j7n/7Tf6qcJwgAG0w0GoWGROYDek84HJbsKrjcD37wg+Hh4S984QvXrl1DX4k67+bm5oGBgfn5+Wg0uri4yKx+9OgRzDy2E3onPnr0iFQNrRqDwaBaiR/zGBsb+0/+k/+E3jG0CaYGGLH4v/f3/t4f//Ef//7v/z4eAzwW+TTWAGFAA4HA8vJyLpe7fv06dcLPnHvWmWDeWgYOQpvYd3AqwlwLBHv34GajMopz1tXVheSh8sCqWdP2o/IOu3wENTjeKkEMfozgbLT1rWqF153VMzoG2mon2MYp4Dlek2u3qJ+2J6YZqATtnNI1W4tpEzV6eC+h0mEmYo9IQ/P9WCzGbk8c1N7eXiqVKCBkHwOTUoNd5P8BB6kvmJ6eZmaXSiUKfJF/xHoC+aGvhMdE9gy2EJ26dnZ2EEvr7e3NZDJra2sdHR3pdLq7u3tubo5NeHNzc2hoiHZcLCFY1TJk2sAjkQg9QaTCCnhHjlENOEimqx2tM0qoqSmR7aCu9qiZnfASkuoZPktyt8XlNsQGFqBFMqw1x4V3sLZ6kZn1efGjIVQw3Uul0vDwMJk03m/jk2j66QGdXjCNj66uLluhyusjZoI3xie8awkeSDhMmy4gL7wIoAZwZNInaFcxsGjOqOSKM3zwwQfJZDKdTn/2s5+dmZmZnp5GCjWTyRwdHdFxlMkWDAY1DehaIG9GsqXMNG05sAPxA86ePfv3//7fR1XVccWkEDk7O/t3/s7faW1tjcfj6XT6zJkzIyMjPKzSbhBPi8Ui3iWFi1tbW+AGXpNqU4tosFimhC2wAmEgEYUgJbCSncNiozYZna+enh42eEUhqnTTDUhATZYNPJb3y+fa/BzPlxJNcWC4E/x3NTyTxyanW06Vgz14+yr8tPnLExSowTK30/upMCljZH1bh8VSc/3wTVjMStwJ+iRT19bWRtM5zt/T06PQjLsEej84OEilUiTB0cqRBk0kEikWi7Ozs/CscUKTyWQsFsvn8+vr66FQaGVlBVI2KCQMEPbYlZWV3d3dcDjc1dUldqDEPWjkSuhUKpUWFhak10pdbyqVWlhYWF5eXlpaAqOUAJNq1ovFIkITsh3kzQmXJI1EUgt4R+wf3je4mySwvW+r3lv8REeDWlNvkkQekIgoP6cCn+aAJN/29vaKxWI6naYlPO/9mA/ySRvZSFNfh+RQBIWJgqLlZKttEfvnZ9Kk9BiyqTPyUfgKTKqurq5gMAjjnvPv7u4SuZfL5XA4fO7cObU7YAJbr1a9h0hu6zUpTaJCksPDw2g0iiL20NDQ0dHR17/+9YGBAW+dsNOaGbemWq0uLy+///77Pp/vlVdemZiYGB0d3dnZwbPhHra2ti5fvvy//+//u80JPXPkJaCon73IO9Eq0uc2yLMu9tHHrijuhRj6qsqxGT8hObxfMdYsWCwX2NtivLu7m+yFzeexSW9ubuJsWcEAQlVHdtVm7Bungo95PHZz8AuszG7N4KVmZzMpUzN29l7Bi9GuI+FLk1y6Z5FoisfjyWRycXHx4OAgm83m8/nR0VGYG2x6nZ2d9+/fD4fDkUiEPU01AtlslhalwCzSTecRoOvR+pb4ZWhoKJFIlMvlq1evplIpnB2qHpTkpcQum822tLTE4/H+/v79/f3BwcGOjg66gTj0HTw1vHufzwf4Trm8nGuYISo2s5PPG3/ZuMmLu/2iQIma57F6L+rGCYarOlKewlGwPOZd4XoQ+WL3CYOoVm1tbQXOOs6dK/fgBUa9Eqn1okbuAeiM2cthJQEUIxOuUjprq8uACCAbYBGAFMB2P/zwQ4Bpi12eOnWqp6cnk8nwvEScKMnAPnY4yAJnADdrDnUikfiTP/kTlkk6nRaEaFliTpWHWPkEqZSPc6H333//Jz/5SSQSOX/+/Gc/+1ksV7lcfuutt771rW+RWaV63sZzNS+k4iAHAXfuH59maWnp5MmTbAAyQY7gwYFBUPf396HeSoPMMqNBVAQuS8RNTrG3+YuTLtrd3Y3FYrZWC0qY7S9ocWArj+PlX//8x9NqNzGBHDdedf22OsYS8mz1naj1fE6XI6bC+Pj4nTt3VlZWtre3y+VyIBCgtieXy6EnSWPm3t5elJVgLIEPgKYReLa2tpZKpa6uLuAIkjw7Ozv9/f337t0rFouYfoqqVbjIvr2zszMzMwOmQZvRSCTi8/lKpRJ5eSi0iUQC+CyZTFYqFRp9tbW1IWmtMWH3ImWERiU6mbxFXHjcfAy3EqGiCgF9ynAIlNBcdPZkxb/1VuwxD5vy9voUloLiZXn/PM61DBCq84R0YJeAuV5iYs3jOD61V3DSmhLrhSjLKmNtnaZIJCJ5Awv6YWGVnGdwqtWqdEFBQvb39+/evTs8PPzw4UMhrWhoLC4uUlK7sbGRTqdnZ2dhsq6vr8ubk98ngrCYnarQwcnt7+/H68KwkqzGXWjA3wdDANAg0MS7Yomtr6+/9dZbFy9e/MpXvnLq1KmPPvro29/+NpAjtEKWrUa4XnmO14hb2XGboszlclTYazYCOMiwtnyMviK/s7W1Rb0PL04FIoqNlFuWIjHsQPvurMNhrS0FXITjAqYtrcDaSflw7PpOQOClk3rlUhtPab75tJeEGoniVWl8baMWsetA9PA4GEpmsypEqtXq2toaRYC8Xd4BpKXp6WnSHZBSsebb29tY56amJmLGTCbT2tpKZ6Curi5acPb29nLm7u7uvr4+VBRI0ZCTSSaTFK/v7u4Wi8VUKgXStLW1RTEOKDkJlrt378Lp6e7uRo2B+6FmJ5vN4u+gYrG2tkbqki2XJD4iOJwQSWuR3hkWmLbIq25vb1upEOQuvUVoTlback4cspE3YfDpDgcft5aa1as9G6fPYdMfM8mpJAQ9IshlqdbD9gSwcpF6OsuX8ipkiuBhP7QFn+LqCrJTBTaALDXlIgup6lXNEBD/xNdGUURkfIBvhxSMaahUKs8///zbb78NKTgUCnV0dMzMzAAu80/cl/39fWR+ESyTuDM3T+yrupLXX3/9l37pl27cuHH//v3t7e3V1VUK5Xt6epio9QBMe2BQiFkJKwlesYNbW1s9PT103pAiOWg4hkLGQQkACcMqacwXKF+Apa73Rf5GsjxwGWdnZ8PhsKSBHJZF28ftOwAzi8WihJeVVFOPNwcrE8nS2w/X8X+xy7iG4juxEqkpYfOw7ClwXWs87Zlr0mprotL2TrxZq8cPv7a2ZoH2eu/YqdSw252oV62trRsbG9VqlZAEfffFxUWITZ2dnXfu3GGdz8zMhMPhdDpNhtPv96vvOPOyVCqFQiFyDmzj+/v7JN+LxeLIyIh647a1tS0tLdE3vVwuK4oMhUISZFCDJSpoh4eHT506BfYiCPXg4ABJYgwrJlgSJaVSKZvNorCugjfUkAmBWeqKSMjM8CzLy8s8hYXVhC2oiaoTu3AA4cnrtBv7pzsss7CmW+1lntREbD7p4fUpYG0S3MTj8bW1NenyOEeDRtr1nlE/q9pFZE216JQIAYJfZETU3peNRF6YLLVUKejSqfomyxKT4iigXzAYHBgYgOwveRCS0nQP2N3d9fv9zz///Pz8PFPauwylTTo4ODg8PDw1NfXgwQNKfL/+9a9HIpHm5uZ79+4NDg6CtzYAqTBbckVrxmpKM2K2JiYm/sv/8r/8X//X/zWfz3N76+vrlgbnAA46J3oXDIU3NcJSZSdraWlZWloaGRlRm0RnAz78uJJQvBFv0yXJYzntePSJxdkd0ylHm1APEoR0gSQPp3oWL8HxmWa6XkawHgfP/vZxrXa1WrUlkvYFC+vwqsHqc9lB1sDGxgbMf1oaDg0NLS4uEvvEYjHKxmKx2IkTJ+bm5lZWVrLZbDQahRmirA4FBcViUcm3lpaWkZGRXC5H1nhmZoY2yc3NzSQGOzs7wTFo7dHS0sIWLaElwlL+6/f76RJ9+vRpFvDMzAwVhvF4HJU1Alt8BJUXagvVQY0yLB+VvKdSKQIFNq2enh6AeGdiCYWQoJpDumATlRSZFyw+5nF8cq43KBNW+InMZeODkyAMgOVC51aNCCxeocl2/EaUzgq3/qnN72Owtra2JiYm5ufnGWRcBF6oqGAiBoDaiclgO/7Y5LB6AlDQSN8Mn8/X19dHopLgOpvN4lZjtj7/+c+/9dZbtqWZ0AlOy96WyWTC4fDMzEyxWESwAe/h4cOHdC9iFVBwUG+Idnd3rWus9yKDJYSHap2Ojo5UKvWHf/iH/+Sf/BPQS5atPYPgYBv7M2jiquoGbOtbPD+Ga3Fxsb+/X4Va9p02f2zEQU1rdnaW7+wk2GoWiNnv2PPgaYFtBoNBpXZV5CnukN3nGke3Xq+5QdbHmyN8PMHK5TK1GM4feAEX3ZZzFnnW6mVOnEJ1nxUmJX4ElV5cXCyXyzRIPjo6OnPmzN27d5k9EOOvX7+OVJXf769UKplM5t69e21tbQB8INFra2udnZ0jIyMbGxtkG1gk5H8CgUChUNjc3FxZWSmXy9PT0yrVBXM8PDx89dVXR0ZGCoXC4uJiqVTq7e0tFot7e3vpdHppaWlsbIy6L/pywdMSlxb3GeKqDQP39/fpFooZYsaDRdpycEAVvBstG8dYi59gy2oFU3w66MNLk7e/8mau8ZhYUV5qwac+5O2ur6+zL8qYOlqPXlUgB8f/RIcGUyQ2CIXsE4SAonYRLAJVkfqjgDiVSi0uLoKfqAQcz9dSgBH3ODw8TKfTly5d2tjY6Ovroy4uEAgAUpOAqVarg4OD7777bqFQkMX0+lxsPMi4o2oAPH3ixIlHjx6dPXsWQJxEbuONjSyOPa2Ws8rN2Se6urooz2lra+vt7f393//9/+P/+D/u3LnjZINlAfhbIRjIm6hyXd+XNI2uApq6uLgI8EIZobVrBx/7Daw+5I2c6kHrWTtxg+ysvmyzGnZGCZvWriy0XQGilwjgdXTqAZXOMvdmWSxP/KfGGhTGamVJoMR61roe7g/4NVMcPX7qEufm5hBL4qkSiUQmk+GvisVib2/v6dOnFxcX6XNBl/FsNos8AlgwTQs5czKZRFLD5/PhRIBR+Hw+wGImOmQMbB/+eGdn5/j4+MbGxvLy8tzcXLlcRk6FvCWMMbITXV1dmUyGZri0ZEYxslwuY2G1YaAauLy8LDEdkip48ZJwsyVPhBfqvYJ6snjfzGyelJ2c7MTOzg4Ex3K5DIqCnpRTDusYVuUkG6xPQXt4QGw2FiWvWa+oeWkzyYqBnEl//EPqdKTXqQxaX18n/NKS89LJLanDy2ateVilMA2UMl1IbLe1tVGRhKvBJARKhk20u7t76tQp4KyFhQX0QOARiYhpb2N7ezsUCpXL5bm5OTq7NzU1LS8vnzp1an5+vqWlJZPJlEolwOKtra10Ov3222/bJL9XjnF0dHRxcREFNJbG2NhYLpcrFAovvvgi5fL8YaVSsXqbGgqbI0VrQQ6slMJslf/o6CheJIPGdvWNb3zjf/wf/0c651r4TrRoIEFy6WDcVjZLll33EwqFcJY7OjoWFxfPnTtHsYXSd0dPXpbUozo7O8PhMIXyzhDJYlqc1paY1Zw8TtAGi1Quue6EjO7u7m5PT49FNe3RgAfCy4XSVvP7DoymnCee4lPlNrsq6qEn9oxSosLKEE7irgqQJXYoFAq0I0JVfXZ2NpfLJRIJlsT8/DxYm7wYdaijuiyTyZBjhE6HZn84HMZtR40PGgkzYGtr6+7du6+99trExMTCwgI1o3Z+cFqw78nJSeYimQocZFxvmB68nt7e3tu3b/f09CwsLOBHix8N+7W9vT0cDlPcodDVdoQB2qMnpP2OkFDSsGQwWlpa1tfXNRVU72SDd2Vvnqmi4D1UL2vFB+qdQRU9z+wa9YkO1hJmmrKjdDrNmsQRE5XVZjLtcfxrWbUZbXLqb4JTz9tnv2fwuUNIohDL+vr6YBbTA9d6pjUFTEKhEDYatv7Dhw8BVZaXl3kuOpcHAoF8Pt/e3o5XUZOmaY0167ZcLuNcx2Kxubk5qE1MYLYNCRvUG5Pbt2/DL7RWTNuDIyXK5xh3rNV/8B/8B//L//K/eO1D423bpvg4p4pclJItFotTU1NjY2P4kd5HaG6YV28gce7VR20wkWoqo9p0uiWPWo6W9WOcoFDz0PtQ9pvWZGujejxp+Utv/ZsepjGXVvGOpG/tQ4ZCoXfeecfn88XjcaJpWnPinOKx7u/vLy8v05qAeJPisZ6envX19aWlJbA/YDgqcSmr6+npefnll3/4wx8eHR09fPgQEC0SiQwPD588eRLp1GQySatGXhULjLwijjkN7tB4QtdJGdRQKBSNRsvl8vXr19fW1vhQ60E7Nkh3pVLRoIuNz9AdHByQ3QZVJ+7e29vz+XzEhvB8QW9shI4onQYT0+DIQh7/8DogTneoehwviTDoO5/Uj655w4RWimRpTIzvViwWoV7UbOJ3zCvW/Nxbm4DA+srKCqaBOUb2GMiyre1xlW8sFhsZGSEkeu+99xxedr3b8Pl84+PjdNJAswzaGaEJCUbKRAOBwPT09Pr6uuXIahrotkFdOjo6KHTkNnAgbNUyEhyNJwmbh00v272wZj9GegnBFVGOuoF/6gy1jWacjZ8QlgXV29u7sLAwNjZmWYxNnzxPo+vaWpjGzmjjo3HkWm8FWYPODmpzGw3O5iWkP1Xk8OKY1nx7lbkplBDdSjlZJyzd399HZw7+XGdnJ8WKkmTc29tbX1/HWokCSG4d4v3W1hZ01OHh4UKhAMJOira9vX12dnZ3dzeTyXz2s5+9fPny9PT0xYsXQR6amprS6fT9+/fxlFVnzEGce/78+TNnzuzt7ilks+IplIFBIUAXDQ0zu3sRYTAsSFlqysqqEjmK+a86T03Zw8PDlZUVEpjY60AgAAxC7OwIFSkf7cz4BpPJWmR4aaxGPa8XcNDnethPJ4BX81CjDWJk6J6UNRcKBdT9GZBfoKC2vTo/6FVyD9hQqcrAuSRRAWstmUz+2Z/9WVNTEzOTDKQjDKSDfEZXV9fMzMwv/dIvCS0RHAmiuLGxgSLN9PR0d3e3yG3Ww5Lp3NzchM0NpNve3r68vJxMJqkn5B3h+zMzG7iZfN/RAKlpcdCQwdehMrOtrW14eFh/Ir/SwnSWb2qDfaegyeEac//ZbNbq4dhbajHntGf2ftO6+TURgprD0vgLdogcUWx7BmcYbYSkRq/Onzg5Qm2Q9g22vvjii7YQxlmQ1hGzTbysrhVTXOLcYuyHw+Hh4WF857m5Odbn9vZ2KpUi3MO1rFQqLAnWA2rC5F6YoJDwW1tbR0ZGYrHY2toa3lA8Hp+enj48PAyHw8iNcofpdHpzc/NHP/rR6OgooKHP5yNLwBVxZL74xS+Wy+UTJ06s5dfy+TyuXEdHB82qQ6HQ6urqzMwM7MPt7e2VJwf+oKIhhNYwsuBZUn2UvaYaFd4IkUcwGCT8JBzB6JPXVhGtIhV8bfEaNfJOPxfvPHPgAu4HPBE/Xf0q60FsNssvUMJ2kPPSve0EtZCFbTGhRAhbpmBWth+ZD6WebL87p+2k49Y1AEm0j+IO4+TaBxwaGqL+BaUBmiSkUinseFtb2/Ly8vz8fGdnJ1I+wk91bw5cQynj2toaARmdRUl9K7xF2ozSweXlZU4lH1NjS8UWclH7+/tnzpwJBoM3b97EgahUKp///OcjkUg4HN7c3GQh4KN4x0GT4cMPPyReFHHCMY6452fPng2HwxTQs2+xyoLB4HvvvSfeuhLgXhDZLhmwWp6LQnz2PMZH8p8tLS3j4+PK2DsocFtbG937AA8dQ8watACancn1eIo1nWINC/tTZ2cnJDFpn3nP47W59s5lQrEPzqKz25jTZoAfanvWlm/gDW20SlVeZQVctDboiIhh6urqYlixm0CloM9ouOAOUw4+NDSEXkEkEgHshtdRKBQSicT4+DgIIM2nMe5zc3O7u7tnzpwhEXT58uVAIDA1NUWxezwehyLW1tbW19dXLpfJ6ft8vg8//NDv92cyGUoVsKpIGMdisd7eXloWQDWhQEZ729HR0crKChsAL9KOjxL6Kl1V2RFdSkVjYmFjuUBpMCXsYdwSKFA9zmyt9eh+R29NL0v+deO/9easP8XhzEtAVa+XJHVpuXufLvh1Dqtmqd3UhikYBbJGqsjY3d3N5/PJZBKKSLVaffDgAcFfJpPJZrMbGxsNNKZl93/yk58sLi6+8cYbv/Zrv3b9+vWZmRlofHghe3t7Z86cuXr1KqQLQEK7IalsuqurC/bn1tYWM1/1clhSFpq6GtmyFO8Ri8WAHJ2Bks2lUolKGYlYWWtI2l8JGM7glVflsHZH0TN0aYgDmhs9PT0jIyOoU6hi9vBnc+m2XsZhJXtLByyBWs5ozcKrmj+rZYE6IBO11PSd7SDYf3obDth7dpL/Wq3O07W+/vrr+oZzbeu1OakDrXlWnd6lLV5PpVKnTp26e/cuJYXqFau6TFLSpE1wB5ivUKlAsbe2tgYHB+fn56nfrVar4+Pj29vb6+vrbLDgGxcuXCiXy/Pz8/QxiD45SqXS3bt329ra6PSRSCQoY1lbWwMuHBwc5NmRJ33ppZcoYSDFNzQ0RMXQgwcPNjc3yR+ixKiZJ0YqZfGaJUQbNjHN1/AscJ2s4wPPDwekr68PGpO6Aor0alveOZ61syRqmnXVKFuWobaTmoeN7GqKitULGO3nktqRNDnChAIEdKCPgf/rTPdnetb1Fo91IEQzJQ7lVCByPT097IikH9Q7ra+vb2BgIJ1Ol0olMnIEcwg9NmgYvbe3VyqVEIepVCq3b99eX18fHh6ORqO0lJMIRmdn55UrV5AfgFilUlvRgbBizEA4RcViEd0ootgvf/nL6O2BiTf2rJHxu3v3rvo6Ov23RFB77rnnmAmk9GGIc8WmpqZbt25JmN87D1VrqkIBG1phNHBT1JaPCfDCCy9AylZ+m0PTErRQu6w+JxRjJ7Df1+ySP+ucsKaah9a4rJYeTfqo9VaB1/QTVdtQxvsdaYvXTAa0vvrqq2J6O9POboaO46yTYqwtC13KjX6/PxaLVSoVULZEIsF0hFsqTTImHPp87e3tQ0NDxWIxHo/zGuLxOCIb3d3dqVQKIkd3d/f8/Hwmk0FfpqWlZWVlJZFIFItFuBmURK6srJw8eVLlod3d3dFoFAVU3Pbu7m48XJjdmHJYXCg9LS8vU2w9OztL2I47oLFC+EbpEUsf1sYoFcTW1lbWm8PDY64j6AV/hpkEmYlZQiwprV4b93gXSU2DpZWjyj2WX4PEiP7WMdb28+MYa3kNumcFzl5XXewXC6M7LYTqXaXe57Id/FN4jqhQuBG8cdU78DkqNB988AHWFhiaci3pNDhjImgVmnCxWETDADZUJpOJxWKa/6lUCoEwn88XjUZROBC5G1CIdEgoFELZplqtLi4ugqTjtaRSqS996UtYOqi0anJf77UWi8WrV6+i+em8aIkI7u7uTkxMkG1CYYKR7OjoGBgY8Pl83//+9/mElKNl+jrG2sqECe5gxUGTlYXa398fGhoiZJGlPjJYsLqnW4EnHfrc8pTtJK/nlDhYlmYOpHVFDDqPU5BccwbaVyCP3gpbO9+3MIiX5dUmnxfwCGOk9vI116Qdd3oYaluzKl/w8x89esSpKGxZW1sD/QiFQtlsNpVKlUqlra0tarh9Pl8qlbpz5w6Ioc/ny+Vya2trwWAwEAhQYkDs2d3dTTVjJBLp6elZXV3Ffc7lcjs7O6lUanh4+MaNG7CqSdrAPZBfQFYHkikuUi6XGxoaWl9fB/ujqVIsFhscHHz06JF1RZVhIF+HC8CHwWBQ/DxL8cF9E90KJSM7mWDFUmqsl6LTMs6U1eE31dyBvWQ+gcvq5GDFIRvoKdrXbYNczTxb7eZ83zthvJ/Yugkbwz2llNYK4R22YgOQpOZdWQjI2eeo5EJsHSlHaBtsmaurqyTWpJbjCEw6CQPtBLx0UGniiY2NjbfffvtXf/VXqQMg4PvRj35EsYxasBMynjhx4sGDBwimo7fDyNP2COiTu2IjB0SKx+PMwAbGGmoHreY0pTkz+Um18hkcHCQ+wAlT7VK5XB4fH0fuQwGfY+a4AWVWLe+NlDKIH4VRYIksxoWFhZMnT6p5kONEHn6cQfH7/TYfy2FbqRwnYWhPawsOrFAdWA2ALfiYepY2OKezlGROVabgPWyVOLRRS6x8bLZoW4UOr1XwqekyOPuPWJNOfzMVL+AdIPk/OzuLZ5fL5Uidkd2Gvko70Xw+Hw6Hp6eno9FoLpfr6uqCpURxCuxL/FzIGwxcPB5Xbc7m5mYikejs7HzxxRdxTpEejkajR0dH6XQaixmJRK5fvx4MBumZBAyXz+cRWiMXRCZBVtXuwBaE0lNLldiaUQ0powTUY4nSzguLRqN0Qwcj0uwJBAKiitfUEa6X4ZFb7WiEOmBO4+MXgh3rVDbCtb/i5lWl9Qs8HPxEH+q5eNdSQMQaQlWiw6dtAufthM1hNTYVZSsDqUTl9773vS9/+csnTpyYmJh49913vcA303h5eXliYiKRSFy/fp1dtl6nmFOnTsHZYBd3YqCaoxGPx8+cOVOpVBYWFpBR29zcRL0a+JEx2dzcRIcPN45hLBaLOC7j4+PXr1+HEuNc0SrjO4I23mw2GX71cCiXy4VCAXkTb8Ks6WNGjTx9uzatE1Nzv3cgDvsrUSScy+EkOSqGzU0/UzBoj5orS/cja1AzrBSS7FW4fCpQSVm2NM+8M9uBbOyTSFLS+ROGnoQMZn1tbQ0zyuZMa5h8Po/G49TUFOH5/v4+XTNgkpLboasLiUeq9WF3wRuRXvjW1lYwGBwfH7906VJHR0c2m/X7/USa8/PzY2NjKysrKI3Nz8+jsefz+RKJBP29qIbP5XLM1J6enrt37w4ODtq5JUOJU2Dfq7h34B4q9sPpwDypUXS9vm0IsSr7J4ZQqVSijQMeVs3DujD2v7bxnXeK1FRO0J//As20k2jynl8ejaMLUfMxn6lbpi87nkfNBQbaqyYv8gZwh0HAeOnigThJHZEHLFUGEx+Px+m6S1vk7e3tH/7wh9SLX7x4EcE829SCCuFgMLi4uDg5OSnZUufxOSCxyNdTxZaX3GmP559//uzZs8zVjY2NUqm0sbExNzd3/fr11SfH5uZmJBLZ3t6en59n/qjQsbu7GzXjYDAIpuHVzNGtah9yfIUGrJ61tTW6i0iJU4c2P0A8saccH8Wx8s/0rHVyR7BeH1LqcZyTyEWzZ9CvMH220N8uAZ6IISXpauGXx3BqPp9HW0M9JWtqL1gxHaedD9+0PHz1PeO/CD+qPI+ZtL6+XigUiH3W1tYQYie3xv6BUV56cvT396fT6VwuNzMzE41GW1tbqSIj/b2xsZHJZFpaWubn52Ox2NLSEt4QJL/Ozs5oNLqwsACsQduXsbGx7u5uRHzu3r3LOVdWVnCo+/v70QkbHBy0L0kCeCwJpz0rDEpvL0rB+ug6SrJKhz0PDa4sPvv0VT0BPbu7uwcHB9VFzDtR9LmVY6VjNzdmCQDHMcTWSXRESxq4b17oo2bSvCaQUpP/9Is6aj6y+FjEnqrYVHVJT08PyWep8TlxvUWWcMDhdNqlq95gVLG/9dZbd+/eLZVK/f39zH97P93d3ZOTk2RrlpeX6T3kNUMcCwsLGxsboVCItaaYr6axtkIFzKvIk6Opqen06dOjo6O06F1ZWSmVSjhD7ARa5pFIZHV1Fe4Wq6Ce0DnRs0KHmhEhDUCoUCc3CGBY06to+rgs3gtcOATEBrypxhUJXq9FfaXt54dHP5Ustue3ZU0ihip/qGHRW3CGTsi7Krp/ylAYHR0lfUF+T261TelYrp/Xc7GpxVAoRH6mq6srmUzOzs7u7e3F43GAYCQ4xsfHqQ8EZaZUBAIjhCEJmUrmcWxsbG1tLZvNDg4O0ndGjeYo80PAFwWy7u5u2EWjo6PLy8uvvPIKBeX0M93e3p6YmIBphxAwV2ltbeVuBwYGyDXNz8/jQ12+fPnevXv5fB6CCjkffGfkE1mfOBcMo6TjlAykSE9iIPBMpZJqWxaRgNXc5W3xRMQQ0BC5Exi4aAP5fvZQZAO/wvJ25bPbTLplajoeq0xzzZjUu/YcsiCf20nlSOHUPKGTLNV4ejF6u8CUSmF+Q4iUhICuSD2hDBZkavqu8o74QfQ+mGTwTVUDLfeFxcy8padzJBKhdJaJzauXPpRs98HBQU9Pz/j4OP0ruI1kMpnJZG7duoWbLE5uLBaTCBGrDC+so6Pji1/8YiqVkooT6Q2VCnsPDaMTKLPo1N8ZdV9qfyAasgkRMUQikb/4i7+goR1jS2ERyCGir9KYZZFqXYitAV0Vp4omkxBg6IutorNmAzxaFpPD8rTUJqJY9hLQeW1jDWYvn5OusDNta2urt7dXlau87pq5StvawpmlenbhxpZQq2F0lAx05sdSG9YFbvAMXl/J3gG/RZhGNiUajfb09AwODhaLxeXlZc68srICTo1QEQgd/GvE10GrJyYmyuUyScixsTGqVNbW1qivzeVy1WqVBgKvvfba9vb2zMxMZ2dnPB7/4IMPQH7h5GWz2bm5OXIvm5ubOCDZbJZ5eXBw8L3vfa9UKsViMb/fv7q6iuQpCqulUun06dMPHjyYnZ3d2NjA33GsJzr6/KDO1raqjRV+eHhohdCQJ2S9WUcbz5fSYcZHb8s6GtQmMPIQSxpwM5zJxOJnMT+TdWTXgDXWcm28l6tHp6vn6Nkw0Iti/8IPLUJLuVUbBJoMUHxIRsSWMKAwJVzL7kNiHGKCSWtLS92mH6SJzDzp6urCrkWj0UKhcHR0BOcaPXcwQ1XYusTbj/vXsFU3NzdTHy/fqyYUUG+SyHaT5Q4Gg7u7u6VSCUq4GoTik+EfoMan+cAWArdKKisS9XTYzZhaXJNwODw/P18oFGKxGDdDtklJ1yPjtEqF1QlrLFqtBJ0Ucb0Evno4G46ULoRDKRwMhjH22jHoDXx5BzGrqdmg7ZzTWiTwaXdzKXI4QbT3ASzIWDM5ZmO94pOjWq2GQqFvf/vbe3t7Q0NDjx49ohsLOwlINPLT7BkEX0Alfr8fohIzUt0OCerxiHGdbt68CZMElIfXn81mKYkkNcSSSKfTgUCgXC5jypPJJKAkO+fu7m4oFBoeHj5x4sStW7cAsk+cOPH222+jQQwTY319XToeLEJNdLXssiEtPTLoV+2VSMW+272QfC/uoSA/5x3LPVcYpWlqYz1dyxYXMMnYtxzJ0wYzqcGUaHA40oAyc1a6r15MKr+j6RdxeBeVNV7QK9nRSa+pcpL4l9fEz6Jpc4Db2kEGUcGZsMlDe12pV0JhIkqjp1I+n0dVSlloJeo1Mmr3jAXH6SZhwyogU/fpRglZOHgBXIJ1RAQ8OjpKN0JtSLorq3/J09kNoyZFh9Ih6gMFxlL4g7aEDQ6aP57wuA4ahJplJoKzarYMtmbdezjdmgjFGFhCXnAhLU8Hjnf6Hjg35m3x4bwCvkbPB+YJr/KxV1jzpmuysr2dGB1SDg5mtVrN5XIffvhhe3s7JR6o+1NWyyQm5F9dXdXOHI1G+VW1WoVyD8dTcCECpKC6Fy5cCIVCCwsL0KLJ1RSLRYjbhUKB/l7hcPjChQvvv/9+Op1ubW29fft2Z2dnf39/c3PziRMnqIqMRqPFYnF1dTWTyYAXp9Npert0dnZSYp5KpVBMpSGZ+hg4QJj2W4dRJO65jeKVgCLu08vD5afOXoPv/EAnCy9FyXmDzuvDl+RQMlpOilM05egSfLo0o+N6NF4hugcHb3HupJ4+mXMcB++2zF9l5KSnqPY9IHUSU9zf30+lUrqK3D1d0XbV4W/lDNmr468AkgAUoEwZj8cLhQL8CudPGBlrSlRkJNv6lOb1s421nnlYVITAAgFYCVlAQCiVSpcvX/7sZz/b3Nz88OFD21BRBajQTKkYwLx4Q0NdVG2h6JiqQYNGTOdVYYlHH+ONbJlag/aF6md13EbvtObbr5fAUC8hIE08JxIbxNYq25bQcc01Us/xVbMR53Nl/iSuogs9Lb7d29tjQ2aAgKt48fLMHfF7kTrRNqIvEQ0HmpqaUNVobm5OJpO/8zu/UywWS6US4CkMvGAwuLS0RDF6Pp8H0e7u7g4EAisrK8iJsLFUq9V0Oq1MN28LWdRyuXz69Om1tbXR0VGEF6Bmo8i6u7sbDoczmcz6+vrExITf70edfXl5eX9/HxLIyMhIKBSCGZ1OpxHJXV9fn5ycHB4eTiaT2Wx2bGxsaWnpypUrLS0tiGRaM1EsFq0BcppfCFmTCI5de9agAzHjmIuDFQgEbDNmvHJiCIISJ6PgxX8VG1rnQoz4QCBgS+TtLmIZmQLTnfVWzzWwuVNbv1Pzy07SCVTRyw+xEpRe8UmHUqm8mXxz/aAmmZSWsLMin0ToQxk3rvHe3h6ZQAJ8Bg0aKIWsLBaNA8tEjy+N+JpZNfUhRD8d5b9YLNbX1xeNRn/yk594E2UWvlS1SKVSiUaj3BgyOPizvHHkJHEGn8mGVGEehWAKuJmN0WiUxUJngHA4PDU1JeCICl4VNAErgW0CCXpni0oQcfLorcq1hG4ToEDL6XnSqMQaJfnyTl8hXg3xNLG76Gre6FPqiVCteN1szIwG28n29jaadJSecjNEM8BlelNIWXB+G9xQWaNlqzG31T12cKhH4SoAxT+lstuVoKaF1lhbW4Dbz3Ag8aUyyrW1NQwi/A1M4eDgYCAQSCQS165dY+35/X5ApXA4PDExsf7kWFhYmJiYoMnsyMjI5cuXX3zxxdu3b4+Pj+dyOXplwf3Y3d0dHh6uVqvJZHJ9ff3VV1/94Q9/eOLECZwCSnKKxWI4HJ6cnORt8V8ebWdn58c//vHg4ODu7u6DBw9aW1sHBgZA3KhT7+vru3fvHhYzmUwCrpHusBkPNRNgd5Vdsyxs8V55XhsKeRkgdiWrXRPvAslZtklH6sXSJ7xRp8pP+JpSi5ocToGMQ8qUcHlNRYXjHI05ZDpqVrs4f+ug89Zk201LfQB0q+rNqn0L0pF1islBwURSUoH3DllCEQ8U0galaDYX6mW+6wByUV9HIJR79+55T+s8voq8YrEYzELF+7LUgtoODw/rZRrt7UGuBbLDFSOpCACNwG88Hocy0NLSsri4KM2Jg4ODaDSq9CAkRTnONT1rO1dx8nw+XzAYVL95epjBc/U/GXmtKSvqYueM5jkAFHLKdv54PRs1RAbqxFuyScjDw8O+vj46Y+iloyTDqSxVpiZx0Eau1n3WDPfWqTAaSM7xXI9rFRGvoWmI1G0020Q5sOKfko+CK03jooODA3TW0VbPZDKhJwejQNOAubk5yHblcpmtHhXKg4OD4eHhycnJQCBAdRbZiUQiMTk5qTAT9//kyZMXL15ks8UXqFQqf/Znf9bT0/PRRx9lMhmEWCcnJ2lmKuVVEh2dnZ30Q9jY2MCa9/b2FgqFmZmZfD4/MTEBvL67uzswMNDa2opvEg6HIVry53rxjLuCIMELVIGS/JEOjnXEnNVrBe30Zd6ujDW6Qmhw461of60JC3ohAl1UW4iXHWxhYjn1TirSywiyl6gJ3jnQs/VxHLqn7sqCAM5zOQkr5wZsCkseLheSnh/leXB1mUho4BGxbWxsBAIByBVYCjj7q6urmGzpBziPqcMmmevZdCi3ZK6ASg4PDy9cuLC1tfXWW295x9ABHlUSQcpLerN2bnR0dKyurq6vr4+NjTU1PPAuY7HY1tYWiUHlG8idbmxsPHr0KBwOU/EQCoUSiQTvC1MrZ1bxpSo7LHmm5tXZRJXDBHCApUZyz/+Eo2Kz3AyXFeqRCcYusfbFtvQyu1UJoZSs2DUqjeEpJKtJYZpdWdJvsOcU5KinxjcXj8POZ+lhODElYhjFYrGvrw9xrjbag7IBKsdCjg4La3ETm4Di8aQdwzum7WwmkxkdHUXLCaIPKmWUBkCb43N6CPj9/nK5DMfe5/MRnheLRUh+09PTTIVHjx7t7++XSiV6Rc/Pz6fTaZTGRkdHr127Bjlasu7IDkQiETr2wlRNJpOUhzU1NfH9nZ2d8fHx9vb2a9euYZdDoRDFOB999FE0GqWcgRwLNCOLEjBWwtHkXIsKiYH2NjAUBKG3TqxN9tJ2PiQm6O7u7unpoXefpoWXafeJDkvIs462CuGs/3ico14RjVXRbPzn6pGmD1VPePyHdYTzsarYZXl5QplUd8AXkKPDxMfjcb2CtbW1QCCAKZTQUlvb4y6mNe9BMgOOZrS12m1tbUjc0CDU5/O9++67fX19+ibzxzIfdGQymZ6eHrozy7O2kBdPGovFRF72jrbMqFAdn88HGEjTO/wD7p+MH/OTOE/V2F1dXYCoAP2KyJuOd8DAkXQtP+/u7q6uruLpNz2BFGwvcznXOgncG9GZUVoGdMYx90ZgrHSss9gj8mM4LZCXIBoV7hFnCzbRw3q1qvW59lcLtTtdpRyYtKuri3Zaj9FpZhLureVmcsfSoLA9aQ4PDmnJjIwR9mt1dXV5eRkJYE5N21mkanZ2dgYGBhgIRJcwXmQPZmdnDw4OPv/5z7///vvhcHhsbOyDDz5gVVQqlZGRkfn5+eHh4d3d3eXlZTzuxcXFw8PD6enpYDB469atpaWlaDQK4WlpaenEiRPNzc2I/z58+PC5556DK7q8vIwkAvU1uVwOoTXgtng8PjExsbm5Sahx584d7pksUE9PD7WFPp8P2iYCNCSIaHSiGIqJRTgmCrnTdJmSS2WTeTHYFJJFhDUWsgC3qsmu83ptjv9rjZ1lWzuTSXaqXs/WT30AB6tfhhdo1s3LuDMDHZV3r9X23p5dljb1z+smw0MMZzvgNDc3F4vFaDRKI1oaH+t9kfTGlUNgHXuUSCRWV1f39/dZ+Shkyd90mObeoKpQKEQikYmJiZmZGTZsKJs7Ozvb29uYY9tWjdiUOGB4eHhjY+PUqVNIruvLFmytZ6k1dNwJ6fSHDx/iS9GMlDRS+ckRCAR6e3slAdjR0YGavNIDDllFc8yp7aoXmfl8vlKpBAQv/Q1Q1itXrrz55psKMamykS5rPp9n7bOKY7FYT08P7iA0XynFy1Zq/OXtHh0dofvxszTXp3erQmJJuaJdpXC5XgMKMbK9nQq8rrSzFvgVWBze99P4HWKckt01g2iGBlBPyKAU1lG8A2pgjRGnbGxsZLNZHBl+NTw8HIvF7t69OzAwgKnt6uoqFApXrlw5e/bs3Nzcu+++S2RXKBRoKspJ/H5/Op0GBEBaDzP62c9+tqmpaXFxMZvN9vb2bm1tUT6DOBn5yfn5ebnSGxsbKPBBvOvr6wuHw/fv3+cPmeW7u7ujo6PEHDCsS6VSOBwG38DzpfaHAJlbArvo6OgANiHmcF6GN4T39k8SFQzDzdQEcqkptuccxyldcdxqVd9wCcVl9ss/v8l+pmctnFdjUhNDeOZhEztKujDg6XS6ra1tY2Ojt7d3fn4eMq9eDSinSl3q0WzY2NTjjRekJJLTndrCTd6DBM/p06f39vbm5ubIsWM7ULOx/XPx6egrvbm5GQqFEFHAOXDwVpwAkj0NusbYAwAQbwMJChqZinGsVgltbW03b9609OfGPO5nHni4iHer1JM3UqlULl++/Ff/6l9FCL6npwfCiRxhwmXsEupOlGWSJpW8otdEsiXws5aq3U1J9Hl7qNvaLs4jnRn7HTGvHOtfc6BqMqGJHohpHtPR8Dp1eZsEs3zhjY0NVVJQiLG787gh/OrqKj4mAo+iu6vLYjQaRRS0Wq1mMpnl5WWKa7e2tugZurKyQj0LGkzlcnlvb0/9g1OpFKh3tVqlFUAul6PQMRqNDg0NDQwM/OhHP+LSiUSCyBSHOpfLEU2EQiEaEdy7dw/16lAohGGiFwGJR9Cr3d3dXC5HmpF6d1L2OGI0nQLDouzNakfwAujp5+1g6aV8cjBNSZySY5FxV0wjbrXesfg9P48x1U1iznBA1Hz90zH2vIfgeAek5gdLFpSQjVy2xjevM9T7JqkR3q/f72eG0N5eRALpI/IW6N/a4NJ0KsFGQ8Kt57raW/XmISFiHR4eTk1NpVIpepxnMpn79+8nEgmr46pmJeFwGE8tk8l86Utf+sY3vgG2g//otciwxBpbautfyw+Nx+NAImtrj1spkVk9OjpiOc/Pzyte8W5FjmfdQKdFn/N0FIhSDKw509PTMzs7+9Zbb33+85/HHGMZcFr5ZvXJkUql6K/NhgqRwavYbt+jUjIODdexyDZzYHPIlpTpvF+bELKovXf+23GwP2irhvn+2K5pN5OOh3Yhmx9wS+MPD3f3Hhvrs2fPlsvlUqkkQ8afs9vjFA8MDOBZx2KxqakpXszMzAxFnN3d3eVyeXR0dGpqqlKpUB5KvW8ul2NzO3Xq1PXr10OhUCAQyOVyRHxLS0vVanVqagpOKHVWbDyAFWfPnL1562ZnZ+f09DTTbm1t7fz58+VymerzarW6v7+/trbW1tZWKBTIIQQCATrFoGUKlzEWi/GM1LjDp4HEAg9EwSbQPz9DLWLEbKsLb+fWQCBA9qZarZLISiaTogBSPaHlUS+j2Bge0YdONIorTSpJhKrGvTD+/3MIrbbPchxuCXOV7wcCAeUVpDfk9/vX1taonXOiH8RkGocv9MeQ18/GTCwFDAK+6WUIOOdMp9Orq6v4sLlcLh6P7+zsVCqVVCrl9/sfPHgg1Ig0F7Vd3d3dL7/88h/+4R8+99xzqJVCEAAthRGsfe44pD0dlKIAAVWr1UKhsLi4SJiP2d3e3i4Wi7FYDJvYQASG50V+55lOvVLloklQscG4JZPJfD5/+fLlrq6uF154wWnvgrwqoe3MzAxuHHkXOlOD07JCnYJGm4i2otCWDYULz3KwSWNx3h35BOehGjyy1xvw/okQ1Mds3QsXLtgSTBVuWQaiMmlikokdBbFJaV/pV+AioVB+dHQ0PDx88+ZNipqoJMS37+npwabv7u4uLS2l0+mFhYX19fXZ2VneAfm0TCZDwQtBQHd3N+3jhoaG9vb2bty40dHRkUwmgagwykg+ptKpzc3Nubm5wcFBYCz1HszlcqFQqFKpUC4Vi8XorUfolEo9/kNM2DvvvINPwZeZu8wJaberKBnXGC1DTLadAXIWiGctf5nwljQmrcgAtW1vOifkdPxKL3It5Je7lS6gojN9WeKFku12iNUylDX3iXqH9fq5uiX52wexeU7uQYRRG2By1FTxdTYncoBsP5TVgdFtb2/TCmB5eRlp0HK5rEVImXUikVDgL9eJCcYbwToz7UnCawFLMMuy98RJIAunA4sPvRp/v6Oj42tf+xoODZMwFAoNDQ1Fo1GKZahd+Bt/429kMhmJBfb09Oi58FekyHb8LB+Pv7+/LwHOjY2NSqVC7y6mN5IgbW1ti4uL165dE6HTvhH14RWrRy1mrGaO/b6yo+yUmgk4T/R/aG1tffToEd0ypd0hyw6Wy2iwxmkbz4dYAxJsTvWA1RKRmbZrVvlDrKLDHK3ZFtEuGTs59TXmmzccdHx/yyB4fA+vvvqqPhXWzmYO79jyrOtRSnnBSjKoJLqvr29paYlxBB0ul8vxeBxQhRJEbHc6nW5vb799+/bVq1enp6fz+Xwul5ufn4eB19raurq6urOzg1OczWYxZKg74rMjlxEMBqenpxcXF8lODA4OTk9PY4XZJ2hxm06nKYhfWlpiDYN5UQnm9/sXFxcplGhtbb1y5QoonkqPcNBYEpRIdXd300+AcEQEAKuUpqMBvKAEUbVaRQWf+J38lQgGNs3ofSMc2vDj8fj58+dXV1cxgg4i4bRxwpuwXZGc26v5c73DKegQduyF7Zxkjs2JOY/pmIaaV1SvPHkY+E0dT4St6CSwtramnix2T4LYiz4GC4ElTTAnA2TFhKlNVd8mZ3xEVLdafRylUon6rOnp6QcPHhwcHKRSqfNPDuloIyFJsmdiYuLcuXO/8Ru/QUGNaqbIbaoYWr0CPmk8RChcLpc3NzdpY0aSxha4d3d3t7e337t3D3YWQ11v15SWqeQ4nMyNvq/0qWUl8QYxzT6fj9bb7e3tY2NjejWhUIgtDVvMKha5xQrGan1ZFqDlbjseVeMOjUpf25Kcmqi0F35RIYzF0K1lqAkVPs3L2/54lt/qRWS8qe2auV2IEL29vdTCZLPZSCSCghI6XiMjI0tLS5QpYum++93v8laot6aFXVNT00cfffTrv/7rmUxmd3f3ypUrAwMDxWLx1KlT165dW11dpf67XC4Tez569AhiBrJN9+7d29vbo5yyubl5YmKiWq2OjY1lMhmSJxKuKxQK9+/fb25uvnDhAms1Ho/Tv4bMu6rDGShcqlAoJMGacDgM6VAdLoCAbQ7Avipnsmrc8IJxvcVJp/m6cxyfqrG4uAjnv2bkpcnkuBuOH/3zSCzZLukOtFoP1XFKH59Zp27/UIoK5BWUkDg8OuzueqzQlM1mCfDV0E9lX9vb24VCAZgLRXyJyTEH4JPociq0EdwvOEX3Y1vC27f26quvAoMsLS2BRk5OTn73u9995ZVXaPM4MzNDHQMGkfjA+y5Ik25sbITDYfzfBgyQBoeQnK2tLSy1w2oXaEateQOquxwLMVmtHpMDZIM0Qg8rlUqi28OCYN/tfJLz7OzsvHTp0sjISDgcZiVCB8CPhocjO/uJfAtLp/MGfzWfUTsxPzvpR28wqpOLINhgTSkE+ak1lmfNfCUuI4ayUbw03rxS13wTO+tILw4PD+/v7+dyObqfbG9vP3z4kP0fo3l0dIRSx5/+6Z9KJ0Hdjrmlra2tTCaD2HQ+nz9x4kQwGESyHW+XznhbW1vRaDSdTs/OzgYCgYGBgYmJie9+97uVSuUzn/kMzunCwsLMkyObzZK0qVQqs7OzV65cuXr16s7OTiwW+8xnPlOpVFg84H3Xrl3L5XLU5Uvfg9hKnbfwdFQdi2fKn9tdvR6GYP9pK5ghrgGb8IOlgtVLK9sMJGyWXC6nptTWfXD4SfqVfq53k5/Us9Y0dYhc1vXQPcu9clw2FZLYOead35yWiMfn86GniN7m1tYWnuzt27eh5bGB2VEl2sWZzefzZJJ9Ph85aqgjtv4AKNxbWM98tmvV650tLCzgR1NngJbvzs7Ohx9++ODBA6TPYdZ2d3fH4/H+/v6hoSEmGC2NGK58Pu/3+6vVajweZ+Ids9uO48fRualUKnHzzGeAQeW62S1+8IMf4E/IMthHE9SuahTSBo5Urw4QIX6FWiyfA9PzgspP6INMlbm5OfQhlBij+RQbnteZ9U5m7xuxlckCRhosMX3TDqZzdQfTsF/z9vOt+Xbsb1tfeeUVK3UvhSfh0eyNpM5FS9KTSD8bgot1w1GwCwQCw8PDOzs7d+7ciUQi5Geh1hcKhe3tbSwvGILSd/hEeEO8476+vvX19TNnzty9e3dra+v06dOPHj0ik04SGWHGvr6+np6eQqGwsrJy48aNXC43Pj4+MDCAj//KK6/E43HcpcnJyUKhQCv0YDB45syZ1JMDkA5SfSAQ2Nra+vGPf4zVg/IF3YfrWmdBe6wtb1PVYmNmhRd2YDYQZmJn9YLq+TLOLLEvW2GXhWIc+qf9E++e751VNqTV4Sx+ez/OQrVRpL269ipZAQtVN8asGXYoOhRb0bUOL6S5uRkG7q1btxYWFpQJFyGSkwjIoupsY2OjUCiADJD17e3tBVCimBsIu6urC9WI3d1dqrEs5c62awKWYSMZHBw8PDy8ffv2jRs3tre3uXNShTTZYtbRkjQUCg0MDHCfkUhEMjK0bGZZcX5Spo4U3HGOtrY2VIjVhhE2sWICOue1tra+/fbbFo9yzmNfjRhl9fArLRb2P8IUYmuJbh5+LOegBuFTU1Owcjc2NtQ8tqbGtNfL9gIUdh5aqp93CThzuyauYL+gfzptf4/p7lg+wk+LnXQ9LWCbdq/Zmkw3UU/UCp2m7u5uVEwfPHjg9/sHBwcXFxfb29uhkeAvw1eTnoAd0P39/aWlpXg8vr+/X6lUsLY3b95ErjoWiwUCAdjfaGTDIyRt/cILLwwMDDQ1NT148GBnZ2dpaQmef2dnJ1XmPBeNpVGEmJubGxsbO3ny5EcffRQMBgmvkKRRwX7NMbU7sy1+OU5nrJq4MwonErbXqB6TM+uFquz59V85EULAaqakf1Ey016jb8+sGLaB3HC9wz5XKBTy+/2BQAAwV1nN9fX1qamphYUFgiS1nav5guz8J9WWz+cXFhYCgcD58+d/4zd+49KlS2Q+GGrYqGCmskHeQhjocUJX4RFzDwINANzW19dpYAQFtqenB0h9c3NTLdilwZ/P5wcHBx0F5GO+EZkzYUekyqVPZPsHHh4ewuRrcH7bYAEA3eZaGtyJVC6UOSOyaft449EtVSqVt99++5d/+Zed6q2ad3Ucz1qfWKFzxxzbVWPHuSYRq+Y/vd88zpt6ylepmbKUOcAwKRPqrPl6bdX5ZzgcpvkxZm54eLhQKAwPD9N6mUjT7/dPTU3R5JAELhPROgVIkvr9/t7eXrgNt27damlpGR0dBe+LRqMnTpy4fft2JpNJJBI7OzuXLl0KBoN4QFtbW6dOnSLlCFHkzJkzwWAwl8sdHR2Fw+H29vbFxUXI3UNDQyQ0Ojs7h4aGmpqalMx85mj+/6j7s9hI0yy9D+caZJCx78EI7mSSuVVWZe3TW0ndo5HcljWABRiQgbnRjQBBVwJkA4ZhwDe+MmDAMHxlA7qRr+yR5ZElj3qmO7t6qrqysnKpXLnvjIWxRzC4kwbzV/X02+8XjGRm1czf/+8iwSQjvuX93ve85zznOc8RVGexg53G3YQCzPJu53tVKE25eZsbcOLXzhlgue1tuixbUZt8n9cqPbeOi4yviobxmMCRxXt7Ld9QA4u363a74bGpemJlZYUGQBKtVZmo0/+ydLXgMmG1S6XS8+fP4/H4Rx99lM1mv/zyS3JZgIFKQZsApf7LfFDFE82aaVwLL4XX7fP5QqEQ6RCpLXq9XtlQKHG4ui6XCzcFgP613otJZUOzie1NXEBzePHlV1dXJe/e/gWZtZcXtbKzXARqzaSFjTfd19enm4EWCflyeXk5Fou1THLquChl98qPtWlM6kRC2p/2ezl+TzlTo6b0qBVyWsNqjr7znlC1pvLl9PR0a2sLTtvCwgL6ISMjI8vLy5FIpFKpUHGE/iTcZ/XEIiYFFF5bWwMpHh8fHxkZ+dWvfjU4OLixsbG6ujoxMUFu5O7du76XBxlkFFMJY6lI3N7exnZHo1F4JpBp9vf3nz59ikLNixcvEKJiciAAphXiHEeAe2YMrRIuGvGWL89J8DQTNRB9nIby8gnG/385zOiEo31TD/NwMjFwD4EmarXa3t7exsYGPqn8NaDSNi0EdUKKLJC63d/fr1ar29vba2trs7OzP/jBD7744gu1SjGVTNSHUA8ophpRWi6X8/v9sVjM6/VubW1VKhWIKMFg8Pbt23/+53/OlMPtrbw8CAQ5ic/nw5lATYFChEu61WaegGEn5mADkw9rRnLgVG1aNre8ioV0tf8wA8ggCAzse6m+oOp2KZGura2ht0UyXzzaV2LBTsT5osGRG2FifabTLbna9rNUg3xRS41XvrLzhzRvnSIOExi1pBgs1RJSB9ls1kk6Qd0J2CscDjebzUAgsLS0dO3aNUq26BqD3w0sRRYIQQDSBaQcKV6IRqOwqjs6Oh49epTL5dLpNLHk4eHh119/vbe391d/9VcTExPklw4ODpLJZGdn56NHj5rNZiwW297e9vv9lAPR/oOcTDabHRsb29raIp+ZTqe3trYKhUKlUkGaisBQKqnFYjGdTtPyhtQlVTkYa9E/+vv7gWho0w4J1Mqt43aZxp2sqSyFDIrQw9d9x9YnTZiF5+JniRxxS1bka6b+LwOJmItBqQ5F7vqMprsTALGgbfMpLJuukyjpAhGeYlTKTQ8PD58/f67UJRWzOicvUa01nYtZ/ZbM0ai9PKrVKukTKsJQGjg6OjItmmpfSVNL+yWTySAsPjs7W6lU8vk89Qc+n+9P/uRP/of/4X9QTBAMBuEdp9Pp7u7u+fl58pzkG6lIKJVK9XodTxMP/ZJTQq9pcnISDx28hSWptuu8L7/f//TpU6fYhfMwy//4LrRIK0a39JhQ81C7GRZU7aVwh8gOBDH8u7i4ODMzI6KBGcco2LK0fy23wPqTEnKqULWGy8SR2ztMLXv+WpIAzs+LvwtR7ebNm1NTU+fm5cMPP7TuVdl2y1tRQGcyQ/iXdIrJ2AV6Hh0dXVtbg/mEYhy8HCbl2NhYLpeTBJLZEMEiC3q9XvTP1tfXoej39/en02mKrI6Pjz/++GMazcRisdnZ2UQisbW1xc2gakakCUuU7rrQ+1EUicViGxsb+/v7w8PDLpfr66+/DgaD4XA4mUyura2RlIfAx8qnBRF2AX8KR0DhnmAsPBRuAHNgznu9mGAwyI3R+k/vVfxcwaBmhk2najlpLvrZKv6W6WGPMaV1TaKVc8q2nKBWJtAKb63fWP9VwVhLlLBNs1f963K5vF4vbVxAA/b29iqVCku6Wq0SsUEUMfVD5DFZdt+CYnBmu3//ODk5YZsPhUIjIyM0foO3p89Qm0r4pWvhjqAUxkSlxVdXV9d77723urq6srKCkgEoIvRNSWCTrmcGitLHRRH8kgPonANS8VbRihoaNBqNZrPJuDFviU5Mdu9nn312GWNtIWmaGxJOcM4oS2tMC+Ts9z1l04Okzw6MHTm5pmap002+qLLX/NlKNjiB6YswOivJ7zx/m+VDZSw0x8HBwWAw+NFHH42NjT179uz58+ffwCDWanSOsnWjJoXT6dnxSqLR6OTkZDabpWEughvqNnT16lXUOUCmRHFzFs6p2zRtASAyAyjDK6IxeT6fDwaDlUoF6Q8MZSwW++qrrzgDl6NAHL+GjWFsbCwQCNRqNVoZ0fgRFufx8XEikfjqq69YZnA20C1BqVUN+jT/1PwCC0s9FfCL03LJGqKtZepKm4/PdaUfb72Olju/841Y/1U7O1VqmFLaJu3HeYaLVqmTGyB6tbO83vyW2a6hTcHLRYeZ2YcNwoZHyOV2uwng4PBC46MtfctbMpWIVTuqqg1ervlFaHw+n+/9998H88XDRT2Og8nGAzLhAVW4XLPZhPRNl5ZUKpXNZtHGId3Ht1wuVyaTAVSheh6yCg5Bb29vMplcXFykW7T5OlrOimazKUV/NVIhxWK21iYHI6kDyhfEXHT2E3gzlNYaUkGystTHL31tfVjTA94ejVuHhoacfQZaGs2LkpDWDy2jt/bA40Vmvc3lrHE4PDzs7+8XRS0UCj19+hQz8rs2MdZtmSe9yIdyCuebH4DV//bbb5O+gwpNf6/Dw8OFhQUy3Tpk6bQetAXdunVrdHQ0k8lks9lIJFIsFtFBl2eB6BLC09L+7+/vx+X55JNPoIUmk0nGolqtxuPxUqk0MDCQy+UePnwYDof7+vq+/vrrn/3sZ263e2Fhob+/f3R0lMZdoOoqpqdPBLs6hFZI1lJ/ZhITfZvLsuU702rR9NKcUKpHseT3i1lLINgsjbHu7fLnt3Jrcta0D7XpJ41tNcE30Qmcn3fm1sVNBHajlFTCmFRv6oQiPJjnFCeE2+Bn3mnLu9UBFJbNZv/yL//S7/dTGzU2Nmbi4CAJQickrSkqcVdXFwVfQ0NDT58+DQaDc3NzpCvkI6PgGolE4DuhZ6Q7r9VqGxsbV65cWVlZEf35ojcFhAJnXH6ommqz1ZmN0FQxRA26OgubWVmTiKJp02b+OBtTWKov0qjRTXYaNop76+zsjEajZMWowms5Wy6T/bM8fTOBZPYTeCXcLIxBjshFFf/O7CUGenBwMJPJfPTRR9VqdWVlJZFIuN3uQqHQfevWLXF4IZnrm+aBY6heXyKN8EOlUjGNtchS0Wg0k8mQOy6VSuzkjUYDimtvb+/U1NTS0hKBm9XZAbSOiPX69eu4luTi8X/J/qkFTHd3dzgczuVyZCBh2qJS7ff7V1dXT09PS6XSW2+9hU4NA3rr1q2NjY1UKhUMBhcWFv7oj/7o008/PT4+HhsbQ5i7UChsbGywf/BcUFAVvbKwWXuWar4Yo9DVJT/Le8KrlWiGuWzEEmGCshmQfL8oQjcn3EVQg+mMW4iHlpa1ryhjY12lzUGDD4gWgjXMnkZWhCHJBfYk61Yt/LqlY8GAMLuI1aLRKHiCx+N58eIF4DJeNkUx5qmksmAuM7OFuXlRUxZZLThURtTV1XXr1q1EIrG+vk5C28S+rMEX8ZxWil6vF/AEJQPFGdj3oaEhuk/wjLu7uwC1nD8SiWxvbz9+/Hh6epqsDAl2LkdpnzIfHR0da2trVHUhaXB6ekoVPs3MmJl8kiVvAmL5fP7Ro0cYa707dXjQM6oEHxN/cnJCYgZDD7isD5tqomZjPG5JJKsOw6vQq4FZy+pDdIgnQiMFkRPLPrb3f535bfO7pl9/ER1QkauFilirUvsNU4jbfvflUS6X+/r66vX60NAQ1U+RSOQbfITvsyG3LLUU/iVjqj+ZNDtzEOPxeDqdLhQKU1NTa2tr6J8lk8mFhQVQPNDDdDqdzWYzmQwye9pOlfC5evXq0dFROBy+d+9eJpMZGhpiSV+9enVxcVH2jlw2pQHk65nZy8vLMKJisRjipV6vVzm0+/fvg603Go1EInFycoJ4iNqP4h1b7+OVtdfm0KkE1tlmTf4F1l+9Wc1LmF21Ls+cvchrkHqtOWmsjJ9T2vG1nHdx0pPJJEybloeZpjaXn6nS0P4BzQY0GDXGuVar0WwUNWfKteC0yYiYYvmqWhb9zuxI8spD8Fez2STtbD6a86n1g4kps8mdnJzkcjmpY+sZ1cnb5/Otr693dXXhufM2sZuhUOjWrVuNRmNycvLhw4eJREKCHkAlc3NzyPODopCQ5LT1ep0oM51Ok9KnmzCbh2nvWPvmFsvvnc+rEBOEnXSRmh07mwEJPScJyW3TXxvQsu+lH2lOSObq4OAgZZ8UZFAZpPDuItzDObFbMpGsT6oy3lpfF0Fq0uPE77QGR1fBF+x/eSBJlEwm6/V6Op1G8wT84BtXUe69OZud88y6hnMNy1kgo/jVV181Gg2XyzU2Nra0tIQYCOdBAWNtbW10dJRiaCsN2tnZGXx5DA8P7+3t/epXvxLonslkzs7OPv/885GREQpPqbKlzmV3d5dWitvb2x6P5/bt2w8ePMAjcLvdk5OTKysre3t7k5OT+XyeVs3ZbJbc/f3798fHx3d3dzc2NghOqZ80n/Eika2Wh9IdKsy7fK2H0AMJ5mmQ23+xTYKuzdX17i7ZysvcrqyEp4ruxIiwwD5TiKplgdlrHSxa1VzQgjoQCEDdUZNMXEs1d77MDnSZjD8EDwCTTCaTTqcBYczPWM1TWsJijHmj0YjFYvLTzTa+JycnmUwGs4Xbri6d6KYSYoK6/PKXv6Q+nhw+lQputzuVSg0NDSEun06nUQbe399/8uTJ0tISMg/UElMhgUAx+mUc0h0E32tZVat4gogHAyfxUjwSyyyafUfh2FQqFewXu1HHxW+fOiPyuhh6YgthMlZjZTN+Ml90ywyQybRT+zflnMzzOBFt3pGKMM3ybzYhdW2GO9fZ2ZnJZOixFY1G6/X606dPedfnRvXjjz823WTTU7aqdzDrmrjSEoFYI+hTKn3SwAWnZrpcu3YtGAwWi8WDgwNKWrq6umZnZ9HkxTBR4N/T03Pt2rXR0dFqtbq7uzswMACT+tmzZzdu3IC4Q3wXiUTK5XI6nUYXuKOjA6GyVCpVLBbn5ubq9frw8LA0VPEX2IqPj4+RUR8aGgKD6+rqGh8fx8FJp9OVSgVxKLPtG7lHRlC2+yJws6WahxPnkjapuQtS2qu3oAD8IsNh/h51NLNbIIe6O5pF8HrjZuZdsdsrMyTmB5BAAkxUc2F9TMiDuZFYDZ6t9KO15ZhxpUYDZi7wF9G3uqyqUPDo6IgCbk5iRhi4P9hEkRfbVH5aD8WH2Y/RsqBKpeWztxxkngXusHgjzDQmRjKZdLvdRAnY63Q6zQzBRrP44/G4IDUUM7q6ulZWVpLJ5OTk5PDwcCAQCIVCvb29iDiSGsUVrdfr29vb9XodCBHBNXhyclnOzs7y+fzKygrpTdGiFXYr7weqSV6Uaby7uyspWiuCFyPl5OSkWq3m83mWPOOJNEpXq3a3At+5f5akWsiaLJeW/qXFObmoR4E5GSwsUe+u5QwRL5lZIR0hmkigb5FKpWZnZycnJ+mcsL+/PzMzc3Z2hvxyqVSiEURfX985BmJp72LvCAxNq42eixLiqlYALTIDE5ZQIpGo1+t4WBCNd3Z2SOglk0lpP62vr2cyGY/Hc+XKleXl5VKpRGW53++/cePG9vb2+vo6DcszmUxfXx+xHs06MUlw7+bm5j744IPPPvuMjB9sPEgd5FLX19efPn1KCy4efmJiYnR0tFAohMPh9fV1SLKFQgGRAXRfCbSphaPZsBotAzkpoNYebukTmS/JXJ+my2ZudSbGgvCsjL4Mrpk+cuapnSbGEns0P6AKDrMY1dkj0XQGnb6w5YlTxyF+rnNhWGiASWd2fv4iWpXp9iKJRxkqLhV84b29PfXeRgaIGG53dxdvdH9/nwZyCn0ExVoCk+admOOpt8+rgfgUi8XATC0xIKuCWaXY0ADom8G+oqHmXeO/84OaAUrUl3tmYtAPGwwwEAhsbm52dXW98847CPJtbW35/X7IV6xcWik1m82xsTGi6v/wH/7DT37yE3ZclgC7KTP/008/BfPRvsJ6ZwCFgCOubQZqNK5ktuzu7uJxs0mz16J4jM1Fgtzr9VJq73rp7Znxn8afNqcQfAXQy5N1godmZZ+Zm2mTP7SILi3Paf1VomM8Plsg4+ByuWjYhggSMYrX6x1+eSwvL6+trdVqNfTvGPNisXhew2V2kLJ6SDsjAu08ztpF5wPIsnPVWCxWr9exvLVaDXqQ3+9fW1sLh8P9/f2PHj0iwmKWDw4OPn/+nDd3dnZGP0aPx7O4uEhIRZOwzs7Oubk5dCx7e3v/1t/6W1988cXh4WE2m+U1w9P68ssvnzx5wvCdnJysr6+7XC4a3Uaj0cHBQVRHKOeNRCI4AkdHR5ubmxS/SSbGsrY4ZcQNUg67qILRXLdWgqXjVQeCRNJJMPkbeiOm4bau2PIwscjvUkr+HY/2CLX1Sad+kKYlVowUBbYPmqasCVAJg4+7oDQaM/YypWjWYRoFOhDRZFbEj4uoYJbdN5kz5sFdmfqr+jyCf9bn6c4RDofVSSAUCn366ac//vGP4b9ix7VyQ6HQ8fGxz+dbWlqic82NGzeWlpbQioJdw0mOj48fP36MB6a0GL6/Jg+rXhGnuauZvA5L/ZESU1qqs44IjAQgHBsPbjXEIM+EorLQYRO3aYMpm7+xluFF/rLpSltG3DkrVMEAp0sbMEHS3t5eo9GAL88DHh0dTU1NIfnJawLCYvC733vvPUJCS61clS9OW6ysKLHM4eEhm7OZq+nu7ga8o8cKoovAhYeHh9evXwdHL5fLIyMj3HcgECgWixQ60stnZ2eHLkcul2t8fBymNrYekISoATza7/cTK9HoSEVxXq/33r17z549gzfS1dVVr9dJIlODDjaEwU0kEnBFDg8PS6XS+Pj41tbW0tISMTJwnioaJDwk/5Toj0+2jKAt6OOi2WOxm/kXr0fm1dlDwEoIW7Gbia9pnWh3kU6IJbtslX6ZP1vtC6ybJ+XQaDR+nyryzbCYrPOW7Rl1OLX3VKBhlmWZw0VZNnxhurvS4lIVcVDowBnY453Rg074SghI6VBligYHB6EMmYZJ64vvql+UFYmbdHvTjsTjcZJm8qAnJiZQ64bEhZcDdKAGZqenp+TVqalhp7cGWfRqui/B70bJj15ZPBT94P/sz/4MD1qrwCTwKLxQCs7KB4q2CNkGLxiFNSAXn8+HjQatVjVvx++/C01gzqZmVSKt8dLbvMGL4jYnJ8R8OucZrH/NHcgJpKAaRtUFUWBfXx8wFO8F7oPL5drY2BgdHWWUSNKe+xlOf8oknVhr9aKkqn4WXZqK3mKxeP369bm5OebuzZs37969OzAwsLy87HK5qEXM5/MkTNhDTk9Py+Xy6enp4uIiheY0psPaRiKRfD7v8/ni8TjA2fDw8NLSUiwWC4VCJycnX3zxRSAQuH37NtXnNGxcWlrCnyLuQAcVcBMf/MaNG/yJ5LjX6w0EAiRkiLDI5Pb19QWDwVKppCnI/KAcDrxld3dXQiItF7mg/4uY+RfxTJCX0sZgGmInG8eajqZQlPWuLQ/ijY+LlBety1lfMRtifZdLm7EgvhXiqCT9MGRQ94hOPB5Pf38/3S1o3XmZkoeWh/mMlLbW6/Xu7m6/3w+gYX5Ym7olJyv8kP+a9SZAYfv7+8qU6g4xypZZgdBCNg/rieiCurpYwo3MJdoEI43N52lCHQwGsS+VSmV+fn5nZwePRxQOk4MsaQQRQ8G7dSFxvWDyqXXROSD7rb5HKBSCyr2/vw8dk4Rkv3EeE1ZuNpvA4qFQKJ/PMyZ+v9/ZBtf0Nix48KLOvy0DX6tDi/Xhi3QxQfDZF+fn57u6uujHQuUHzmK9Xne5XNFo9Nq1a1999RVAVjQaBart/uijj8R418VQCOHsJq/QTBBpnWcyGTllCjkHBwfHx8dh0TUajf7+flqSd3Z20lLr5OQkFovVajXcWB7j5ORkdnaW1UXaBGCL1J/691DzUq1Wa7VaoVAYHR2t1+s+n29hYSGXy1Wr1XA4DMG5q6vrt7/97enpKZpNTHfUh9mKmVJXr16dnJzs6Ogol8s8MpnZ4+Pjra2t9fV1Md6ImjVH1aTq9PSU8BBARmQ7yy8QOcHsMNBy2VuTw4r6hYRYM8zysrXVm8k0EVSYcEQD8HAtBr1Z2mDejzUfrI1HsCybmfktxko9xYUgXWS52gQoTo+eLBM+5uTk5NTU1D/5J//kJz/5yeeff45+IeQ8ZKOF/AJ2S9oNh9Q5sNZqtESjZGGV0YlEIl1dXeFwmGSPFEtMQ0ZO0jyz01nj3UGCUptmMNzJyUlOODg4qKwpGALoPC62CXKaL8L5XIVCwefzdXZ2vnjxwuv15vP57e1tiiHJ/fzqV79CZM28W0UMGFPNVVAaggBIk5hgQn4013DFkskk2BFXF5IraB7w+uT388+8OErM9/f36/U6ritfIZNMab6pT2CGjC3nlabcRYtRs8LqkGcptXIG1cRR8HHjxg08xVwuh6mkhz0d5njXy8vLSO3PzMysra2FQqFyuUzfy/OhVKTD62QUnF2IzOVB3MEsV70TPqPEdBKJxPHx8cLCAgvpBz/4ATUpv/rVryjIQTV8cHDw8PAQQ9xsNp8+fXp0dDQ8PNxoNOjdNTo6Sr6RhsqlUoksKs44IRvcj5/+9KdffPFFZ2cnjLFUKnX37l32alMmW6INerXr6+tnZ2fRaLRYLO7u7rIweGRnqZtQIMuESdwSLwDkBBEoWUk5xU5GRxtKiQVAWeGzkzuhy+k9theOkQZAx3c4TO4z/0ov1Ly3luHna7nVVhWPlWnkCAQCw8PD5MEuqrCFUETETX2TNrA3HgruhAC/t7eX7JxWGZ+R0WmZJHAiZngzx8fHsVhsd3fX5/Phi8AdQrN7YGAA4pp6LrdXjnZecWBggAbWx8fHq6ursVjM7/cfHR0tLy/PzMwUCoXV1VU6zFmVU9rYeBy5KXpZZKdUzYtb4Pf7g8GgEqexWOzg4GBnZycYDJIfkgcJ2Wz/5RarsID/kpoD+eRnXDHiAzgkLDonrGRZZOsH09aZfAolkyXBxn+lTmWVLODyer3eoaEh2l3t7++nUqmTk5PV1VUYLLlc7vr16+FwOBQKuVyu2dnZYrH45Zdfbmxs3L59m+iQ8TkH8sWkkcwbUbNaw1izR/GaTLnJ9yQ54HK5EK5jpOjKnM1mc7kcCb1YLPbFF1/gIwwODuIOw8Yjm0/kFY1GV1dXd3Z2EJ7u7Ox8++23BwYGUHuguRElM2ifJpPJWq2GjCT6CX6/v16vs3hMWJOdhlVEtFUqlTo7O8/DjZfyC11dXalUamVlhdyU9S4ZNMqxLrke1DNebaT1AXNsX1luIytsntxpvKwoTyc3q+/UrhAQX69Vr/jNso6cWfQhc2MzG5Y7AZyWmR/z5/YYhenkhkIhiq3UaLGNCcOhNtnNr3uYbwFjPTg4CGXCpAnr/QK2vvKEKn9H6oCmSMfHx6Ojo+yvSNnQNg8EAP7fK8dKh6hKwpqvXbv29OnTSqXi9XoRL+vt7f1//p//R33KW57ZSTrQHkmKkuwulXhAHErAbm1t9fX1DQ8Pk2BUJY50r45eEkVAoq38uYRNSIfiGhK74IaamNJF0ZK1wZiftPxUxSWKVKQPyn9VMkJwHwgE0Nm/f/9+V1cXVIvu7m46xpVKJfQalWOAdTM1NbWwsAB1gtjifCKhB8ZCJZwB8AL8Ni2CBf3IDRFWxeCKaQQsgECS3+///PPPJycnYUpRJPajH/0IaWl88J2dnUqlggrP2toaBSm1Ws3j8UQikUKhQJQXDAZfvHhBAz2ueHR0VCwWIYf29/ffvn1bmWV8gY6Ojng8rjZO8kq0VaL+w1ZcLpdfvHhx48aNjo6Ou3fvkqRyDoKTJIfLQDyovkpm52xrBlhesIkgt2ypJYqu2fOBw2oNYZlsc0+1kpAYaw54Yx2XOGRuWv7VSpSZTHOl4KyvOJ/oOx61Wu309HR1dfXZs2cYa/mDukm4mDDGcFmgi7Q01q/bnUfqvhgRExVleasV7GXOCdyBtQL0ODs7W1xcpBcdKxnJRh7E5I+2PwjVQRIikUij0Tg5ORkZGWk0GqD8x8fHL168AD8lkylGZsv6ETPzzJOSvSeLgDgPAYE2IfnLNHLCe8NeyxB3v/wMRHho5pRlDAwMSKaChcC3MOsQKpwMnPaH82MWRofdA81zOhxmk5DJyUlYetvb24uLi4FAAIw0HA5PTk4uLy/jVmazWb/fD/jDDKSgaW5ubm9vLxKJbG1tpVKp3yUYdX/UX0GxxGoLkNJxcHAAa7JYLEo9gNiEvTeVShFSeTyeYrGYSCQge6AcdvPmzcePHwNrMKBTU1Po11Sr1WAwiMZpNpulq1apVHr//fc3NzcLhcLdu3cPDw89Hg+PVCqVgLyB5wcHB7e2trxe79jY2O7urtfrlfCIKprUlVViOp2dnaOjoxQ9dnZ2zs7OQmIJBALHx8eqnxall9ksDgznFI/StHqmQK1stFluayUJZc6czASLN2oGPRKaMKM8q12W0gyWsZb+KtQdk55hXreNv29tYOZdgdGbMLSpEMINUyCHGHTLDLt5Tmeqx7nSMNZHR0fr6+sej4e4zWRi4I4wadG2Bbnq7e2FoqtGEwBlZpPvi0B8jTMzpNFoRCIRSgFJmJPxp+hGGQs8Bgsakr4jkxNnSKRdujyHw2Gfzzc8PIzeNKkdGGxk6sxX4wTczCQBjjMDQrTa0dFBs1M4IVtbWywH8UD01LhWcBUgcYFHQ/TCg47FYpSohEIhcCHMnGw6Hh6+l5jp7EmsUDbR7m/7Vek2eDUMsqJAoA+K/nmW3d1ds7KMm5fovMafQTYXF/9Fnoj5wD0AApu7L/q0TB6+0tPTMz09TXSFaaJYGsyWltz7+/vNZhOdRUJ/qAp4EnB7GIH19fXr169/c39mZG12U2UbNJn8HNQTM2pmdE8ah1osj8ezuro6PDyMoKj0RSnLIZrr6Oi4ffv248eP7927FwqFMpnMxMREpVJ59uxZPB7PZrPYynA4jLzv4OAgHsTMzAyKTj6fb3NzM5lMIsgHyZQ6dYAUUDCp/TpxBpZQJpOJx+P9/f35fF6qTC6XyyxL4fB4PJSHSWvYslAXGTWtE1P7wmk9W37LfDVWbYUOvQsnadTsrsTZsKRMEYkumfn6ljivztYmA9PyMDuiWu4knggXvaSn+cqjVCotLi6S2dbjW7UtWE/WFf9aOSgTM7nMw5oxE+BsKBQCWqViuNFoqE8bYwhsSCJOXRlFW7ZIyjp5NBqFeAuYpjaS7atbL5pRkD3AT7a2tigatJwz1YxYbgF+CXhpvV6XsgL2msQguVbsFGI7wn9akuGw18qm4H0ffxseyQ2XoQRtgy5srm6KM0KhEDbUekfwgszfm1GgMGEmCY2MURxktlCHof2GV8DTYX8o2igWi93d3bA+RkdH79+/XyqV1tfXR0ZGAoFAJpNhWMwaDtbp6ekpDVuEAp1LLZopAt20ADXstUm/V9oNWhKl4Zhd3jeJtVqtxgTd2toCQgL4x0xUKhUKZLu7u+negoBqKBSqVqtgc1TrRiIRatnhVk9MTPT39z99+vTw8DCdTo+OjhJbPXjwIBQKDQ8PC25eWFig4vn09FRZZk0IXEgOxPnoBknLm8HBwbW1NdAM2plbCAY62hctY6cz2NKCm4iYII5XrjHn11t+wPo9mTSrtB0vTAeRu3qptOcRXvJQhZtZgak9gy2HAgHSYt/FWJsXouGT6U+ZHV44ACgwTMJhqKZRXqElhOVMaVq/xwoXCoVgMCi4A+/ETFTwe2lM4p+2B7JJN3HDRLH8zHlMn9o8nBx585DeHqGwgk5rszfVJlpCH0C3gBI+nw+HFx4Xni/YIP6jU4zQdGVUfU5CRQX0x99KwlrhCGczG5hgxAkX5Bpbr14WWewOyefSqJ4tHAIMe5jZZFlt4bq7u6PRKE6x6q1OTk6g05yensbjcWRqJicnHz9+7PF4kKtuNpsrKyvU2crTx9gidUcWjdV63lRInrWwGMkrKx4nVMFY8xs8a6BnjSmwWnd3NyqUlKskEolSqQSSAKpQrVYBLrDXY2NjDx48UL967hhhwC+//DKdTj958iSTySSTSeoVh4eHi8UifJfJyUl2DvDB58+f0+Xo66+/TqfTkUjkypUrCwsLKApdhDmmUqn+/n560LD9UA/a19dXqVQItXi1ktphClLueHko01JAtabpm5nFi6BJ5++dTQbIAUiTU6GSCXB3fE+HWZps3QNzAx/T/MpFducyniNcXYLQi6BnrIDab0JaQAQDo9BGPOiVD4smcKFQwPj29fXxA0C5QvvBwUHplryyKTOGFUNQrVaHh4d3d3dDoZCZcGqJeFj3Zm3zqoeiCIWqrkaj0XL2Ok/IkmGHg2VIA0n1ZqrVaiggaqapI/ZFKTHdvDTmdAibtl6lE6ln2ZbLZTid+r3H4yF2sToi+Xw+atxdLtfMzAwbORaZbQZoAl8b/5oGysxeKTFRrrG/v7+ysrK9vQ0vLhKJ5HK5UCg0MTGxsbHBegQPoMTa5/P5/X7u33RomDmIQp/rdeilWkxMgXoqVoT7DHUU8wfnyVSTIqry+XzZbJYNbXNzE5pzrVajjOrw8JBpSrG4eK+8gIGBAfofDg8Pb29vU6JCUaZYzPz77NkzdMX8fv/bb7/9+PHjlZWV6elputgdHByMjIzkcjlWLIOIVYLmeXR0dOPGjXA4zGumUobWfDRjJNCDpgLJj3uAyGlqNCsgsjw48efN1nDOqidzglrzzyI16wU5DURL11sL0oycOL8SyOI7W+xsp6tuInoX7S6aRaYGaZvuNmBf6+vralDyBpuEVcRI3sbMApnhiymFykrweDz4PjhoTBjRzqDGO+sgnJbRLHdCZdfsbs7qwF/j5AjBmxq8zjdLwg2eciKRIN4H7uOX7EkyiFbFh9WhqqVl9Hq9akaDsjZJS0hyrHHmP0RbNUgiwpaZVutLnpcsDuNAgpE9ALsve80LQtIH+VZTcUWSTyeGUKK4fVYwgfwT4XVPT0+hUIhEIrBvdR6ycdywnH3o3hTXEAeoVIrNDKtFRS7zWSKC5lRnpddqtc7OTnTl8i8PwTKIPMfj8dHRUZfLdeXKlWAwePXq1XA4TH0/7gIdmdnV8AvPjbU5XiIeKf/DTGKS4YXxS1HQmetqLicOH8V+JAAxfHClqTq5devWX/3VX52cnExNTe3u7o6MjMD2LxQKPT090Wh0d3c3k8lQ/Tg0NFQulycmJp49e9bV1fVXf/VXKE2vra253e4PPvigp6dnbW0NX7ujo6NQKIiyQ7OlJ0+eUHYouNDn85E1crvdGxsb8Xi8UCgwTQGpBcqrpYCTsSBVHZPF2OawiHTO45VnuCRb4zKH5NwIANmSdRXdidyu736YoIFZqk6lK2GKbkChgBMYcW5L8pF1fikEtdxReF6TwgzuCdYHxImv1PF9H1j8/f19cEIRZ/WwrwSCpqamqGVVk0kedmRkhA+YuK2lf9+mT7w2GHwjtDUIqQGpsE2lUonlwPBiK4F0Qa5NOT0O7vP09NTEHsXIlutAHth8j8KgLTel66WfIXAc4gAHfj07AbEakbFIa+j2kJRiHiIbOzU1RfUKtEvAJWoj8Qu5DVN9E9FBzTHdZG9v7+bmZl9fXywWW1hYIKHa39//wQcf7O7uLi0tYb4jkQhKoslkEs/1+PgYoYLd3d1yuYwGGTAaS/V3SZWWr/AitiZ7KTbRosGDSYFBkyecnp6GOwE3xe/3z8/Pj42NLS8vI77l8Xg2NjYmJybdL4+NjQ1kCqDQ/frXv+7q6kKeKZVKCVRhPu3t7dFlvFwuv//++48fP242m3CYGo1GLpfz+XwTExNbW1ssQlytvr4+7DJ+wdraGkxwMuOcVuqapvh6y6BSfRedKjNteGkm5aNlQWOb45JowGUOZXLa1+O82cG0bhmem/gbHQsxB5YhNofORBitq+A9KcgDddXktBIzysQolUoxtww9r8NsL9dSY9OySm3eoOwU1Fp+RsqGaSZtYfNJnaelDPjWrVsw2LDyvKZGo1Eul5VM8/l8NEQ30Xyll9qgZxTXqOSN4kD9jGcmeWRSbfwgzRYTJceNMwWd5fWj+UOlhYIq8zATaZ2XmO2MA1ssCUCXy6WYGPfLVJJSvm1wcLBSqWB5MY79/f0oW+EpFgoFgnuTxSQ2t0Xc6uzsvH79+v379/1+fyqVWltbY8RCoRAph4ODA7pcgRIDtlCcxdwA+q/VapQN4q2f70Mtp11LG23+Bmo6I27JRao9GGJRqCMh9FEul6+8PO7cuZPL5XhPpZeH1+tdXlnu6upaXV0NBAL7+/tXr14FyiAkgbTY19c3MzNzcHCwurpKHRTcnampqWKxuLm5yecnJia2t7eDweD9+/drtRqKkRQxAkLRb6xarY6PjyOqFwqFiO8QvtnY2BD5rA3H1uyg+MbHdy8dfGPDzcxjmYmJaH3gu9zbZc7G8CJYqntwqsJb/HGtEBEGTMxXAhTWq9HZMHYaeT6J34TXptD7zWpkrINOntIZFmPSNHCmjJrzqTkoQUilUl1dXQ8ePMDvc7lcPp+vWq2CUUiJ//DwMBgM4qZB0+rv70fxRg1wrWuhV0eEQTCqAn3aeogXQAN1iIOqQJEdNxEtEn0igAvNoJwN3bRwOAyCAUCh+1F83/n7XFgI7OqJqmtRFUEyRsk2MoQgzrwCRcM85v/9f//fH3zwAawkvuVyubLZLLtjPB4/PDwUwCL/Q/RKyz+jHorCxUAgsLGx0d3dnUqlGMxoNIqGOIKuiD/7fD668zCStVqNUmrCJm77PA0A05MJJBa6qtpMOQsNPVC6dJxNY4F7EgwGqS45OzsDzucFE11Wq9ViscjQZDIZr9cLIQRNFkoTcc8PDw+/+uqrWCzG+dHhrtfrMzMzXq+X//IiK5XK/v4+HceB9tn3YP5VKhUusbu7i2I1DJPBwUGv1ws2xNRhp9na2qKKd35+3ufzSaFNeHRL7NVZsaqFJ/jFYo+Zzp01Hc216qTiXVR25VRebvmz6SoqILUO62HNGdmy0YnZQ6elvCRt95xDJysmaPsiAETcZDMuIYkP9Z6Pya1mH2JHZw3I7yNawh+06mVIjPOzWfxmVUY4i4/MZlccLBOkV/QBE7cxy5edr0DX5bTY2WAwuLS0RGeMcrns8XhSqdSVK1ekFk3DPPi8a2trR0dHs7OzYAJoctL0I5lMHh8fFwoF9ZEaGRmJxWIwEJD/pn4dq40uMfJ7ujG1djSfQitFqREROeQAYblQ1q5Wq7lcDswdOpnZAVnfOvs2DkOoS0lgnZN/8V6xejDKdIfCVTSxubFqtVoqlSjdIHmI1C1tTMbHx6nNNht6mciSiaqLjDAxMdFoNK5cuVIoFN577z2v1yv+WCQSoTqGRhBsnNDDyUg3Go21tbVqtYrv4vV6b9y4MTU11f3222+LWyNXRS2NuTklRqUtqwGt1Wom5E1w0d3djfZ/PB4Ph8MPHz5Ebgnburq6Gg6HK5UKPQ/JFiaTyZ2dHXrHxeNxcL10Og3688knnxwfHweDQaqhwJonJiYKhUKxWKSaw+12+3y+mzdvzs/PA0VtbGzAA4GUhgLJlStXstnswcFBOp3GT3G73dBUPB7PzMxMLBbLZDLURp6dnYGQtDGR8lAuWm8Ci6i/cC7FluVS5mEmGC2FDY18yxIS/Sx3wBJyMq+iVWRxwC1h9TahgPPq3BiLxCz3t7py6CvigZmdSa3DGjrtPeoBz6lInBB+7r48pActbFGS6y21J01i5Suf1PlC5QAp82z6QPqZV2Omv5yP6ff7BwcHEZvEcP/sZz8bHBycnp4eHR3t7++fnp5GJxJoRXy1aDQKzZmVS1FfKBRKJBKFQgGtykgkwvJ/8uTJ3NxcKBRSlWxnZ+fKysr8/DwVQ5ghs70cRRiSFTPv33yzsmsEB36/H/IuWQqyu5RNqLGGwAqgyKOXL9cUDIFFp5IlUfg1pGh5fgMgGMJn5ssiIAMwmZqakmYhlaIStMBfxga23FlVJUDU0tHRgdBQX1/f+Pi4x+PZ29uj3A9FJzKxbrebHZ0D3GNnZ2djYyOfz5N8jkQiN2/evHXr1vl+6VzPTprwZfIeHDCaySaPjo5CWIYTIw7m2dkZWuYQormhQqFQrVZjsRgeAXlnGqK73e7f/OY3kK/Nay0sLOC57OzsqPnQ8+fP1eacAlYYi4jBl8vl9fV1cD2gD7rMFAqFvb099JcXFxePjo7oIYK6Uy6XA11pqd/kLOy2HC4eh1ZDZs7KUmhqaesVBnb89R8tPeXvcmlTctqaP6addV7xMpPNZKnLucZ2SMERN5ly1v7+/uXlc5ztDR7E6VNftD3LucYxFEPf3Kg4TAJ4+6urJRtgETFELBZjGgMFmLZSaLt5UUvIDLRUlpGedt3d3dVq9de//vWtW7dorQ0DUrBJy2eXXJwAAZ1Wj4a3pD0JlSU18YE0QrG0MlskUdSa4PilVyuLLE2brq4urU39RuMGWquEoVIaZhIb7xPQQ8fAwACJNyB7Gge2QTuhzcjNzWQy+JpUDzAJh4aGqFekz4lYHyLd1ev1SqWytraWy+WgSo+MjMzOzg4PD6Mv+DuZcBnrlpyty3RbgIPMqwWzr9frdH67c+fO7u7u6Ojo9vY2dQfi/2OFv/rqK1KfhAmjo6NKN9dqtXfffffhw4f0Ijo5OQmHwzS0VZVNZ2dnsVjEE2czTKVSjx8/9vv9Kysr+XyeLdTr9XZ1dY2MjAwNDT18+DASiRweHmoHm5+fpz8FAcv09PTJyQllMnol1vM64QXnmAgfNEFVp4rCK4/vkfV8yfM7C81fabut3H37s333u9WF1BKe7CL9GGFN7O7uTk9P/6f/6X/6P/6P/6OzUT3H63aOb4NayK/EiFykGN7SRrf8pUITJYTS6TRkJ5w4M4kqAMFC2y0dErXRgPhMyB8Oh6PRKBYKDAGJNEsk0hI20sklmWS1P8dqE1zySwwWaeHT01McKTxl+dEA+uxDrPqDlzg4lYeU1RBAC3+wmHw0+urq6oJoYfn+eha+C9hNlKAwgqPZbPp8vvaiZuQh1Qe8t7f3t7/97bvvvgtDJpFIiPoMvVqtPjs7O2nNjpzR3t5eJpNpNBoDAwPxePwHP/jBxMQExKHz3ejDDz9UZoM92VQF4nRkKkH3kAvgOYkgBIMI8qdW5fnz5/39/dvb20tLS2dnZ4lEgmZa/f39YFUg2vV6/csvvywUCpRdsrtms9mzs7N0On3jxo2VlZVyuQzlgyzl5ubm8PBw+eURj8crlcrg4GA6nRY5msio9vKIxWJQp3El8DhAzMlTr66uUoPElo437Xa7y+UyVflQ0a3SFSaQ+frNamOmGlGP5q5Y/U5LbYW9mkkW29/6mOkpXKT+rImrnVgURituNStizOe9vHk1OQCq0TDjU+vMWswW/dE5Ji3xEKu+w1QKlLQ0BAak6YC/nEp75hDpMHEJZ3G/9iQT2TBp3fgZSmhbEboT4dFpLTvIDVCiBUeNOl5MrUxVKpXCcRYKZEUnF/nv5uSEktvd3U3nPCDpjY2NFy9eeDyeRqOhXkUCx3C8Tk9PKTugIK7RaOBrQ8cEYmZtginr6wwL61qdZHHqMZ0A0PhkR992NLaKzq1TmRMMYg/GWlR6PThs8ZOTEwpbPv74YxqYACXxKmHvjI6OAt9TdC6lFBXXcCcrKyuw9La2tlArRBsOo9ff34/kXjabZS+s1+s9PT0LCwubm5tra2t4t3jfMzMzP/7xj4eGhhqNBkI331D3nJ6FU1FefEk1rzMVXXWAOjWbzWAwyM15PB5Kw7PZ7MzMTLlcbjabhULB4/FEo9H19fWVlRVAYQ036RFEGiuVSiqVAuhA6iEcDns8HpQWPB5PLpdjhwBy0Qzu7e2dmppaWVmhAKdSqRSLRfZzApxarZZKpUZGRuiLQTcZALuFhYVoNJpKpQ4PD1FMNbFR6UnixJmUZE0F3hmtcC6yaybIcBnPrg07+yJo1TraX0XeEP9VSqdN+7GWN3OREulfU2TQEos0u6dDwP/yyy+r1WrLokSnaTP/xA+WNBIG16rIMEFnUSE5uXI8bxZbmOpduJ/KOipr/V2Gl7lNQdDBwcHQ0JDf78/lckxva9BMbUjpqlMnIT1VqrRRcAPuAJwEnGWRmkWSOOb8lbQbuVCzU+DJt/i1+pFKCavliJkpB5BoUCPt0MpDchQKhVAoJKcKTgiv8uDggJJsdIpUua3LeTyecrmcSCQeP348MjICo6bZbK6urt6+fZsWxsFgEKbKtWvX7t69293dvbW1RSPD3d3darVK95yJiYnr169fvXqVzKdAnvN92hQV0nNafZJMJIhREKHdSY2CQr+yspJIJNCXovW4z+ebmZnp7u6+c+cO9TKwxyEVsRXLl4SQV6lUPv/88z/+4z+GNE4tDHaWBrXcw9WrV0Giocv09fWVy2WKO3t6zjVg5+fnkSvJZrNoj5BVL5fLy8vLn3zySTqdpttpIpEAI+MzTkFxCYriZZAnkf9l6m/QaRQo0JkSNKnNluNs0ki+R+acU8jCeuNmRZnFRrr8bTiHS75kS2vScv94rcuZZaJiKJn2i2QGbsRFqULnY1pgoPUZTQMnNogf8/1SHnUQ4+IPsgdI19D5UJc5lBQBDyFXT2t2ulpTuKHgUgcAOjVEEO+woXjECITRxBZZTSwR8AVG2dwgzWFUpgoCPqa/+1vMB3uN06oIps1TMyuQBoSSyFanuSEpj42NjRs3bkiaCv1VrkuZaC6Xg0gD2cEqy8LuDQ8PY83IBh8cHGxsbECFQNqpVCqVy2XqRXDGAQAghMzOzr733nvpdBq6Ifw0PVcLXTHZDmWEWbo8pPKEYsyY1AgSg7dv367X65ubm2ySyWTy3r17yWTyN7/5TbFY7OjomJ2d7ezs/Prrrzs7O2lIASijUyGtS2agWCyiRt3T09NoNEZGRur1+ltvvbWwsAAx8+HDh263O5/PQ4hRpcDg4OD6+vr8/Hw2m6VOlF29t7d3Z2cnn89TPOn/9mg0Gi9evLh169be3h7cIPW1NElphGPa9p2txBkZSmzNGWm+XSnjyD9q+SIusnFvBqdaCXoTS9G/ZleE17XU5rda/v4iNuH3dbS8dDAYpPzXadTMDl5W0wazQp1f8q7xTsSLsIy1/M3v8aEs2QCMoOaG8rGmN3D5k8s/wGYx7QE3+vr6gBMta6hhwf5SljwwMEAXR5jXkBxYntlsFooBgAMvAj058xnJXbFJkNvXo0lczPWtvcYySGmr/e7In6gXRTJTISPQH05nPp/X7os3DRKCCFcgEAAhgRxiNTWlaxVwB6se3Njj8eBfp1KppaWl7u7ujY2N/f39R48evf/++z/96U/v3r37f/6f/+fZ2dno6OjVq1fff/99qfIqlcq7Pl+bZoGAuYDlKnL3Ip8zR30+HybP8qTwFqvV6tDQkMvlWlhYQLjO7/d3d3ePjY0Fg0Gfz9fV1bW9vQ2zcm9vDxDcTMJAV0KeqVKpTE1NITYyNjaGKMHq6iob+OHh4TvvvLO8vEwHip2dnf39/Wg06vV6Efnt7+8PBAK40mLgQs/c3d1ll6ZRBcUyJ8fnWYJKpUIzMJBrgkGpcOFH0OORv1oC0CIMgXCxxiiz1DDWajV+TxLMab/MLohO2oBFH7TErM2PqXJE/xXR1VQuJTUPV8eUjrR41i39TT2UOXk4CFSVZTUFdMxY9aI15vzZySh3IgzmZwjdrE9KX0JkFWY4oLalEKTXavaRwYRJLFBxleZDmyxry4e9qEkNg0YcCTlXmv2sBdZaxxsdeJFHR0foQODiUdQHPS4ajYomb5oIaAz03GKQXS4XzazZ7+m3R2IACp3mAKk2FVIDebOsKNvRGyT1p1Kas5ejiv9OlUpLrWBnIhfa8dHREYCqRSFFgS6TyfAqSTOQeCSbBbCZTCZZ/ipoopNfs9mUYIv5ecAQ6kXm5+dTqdT+/r6r99wWb2xsLCwsDA0N/fSnP202m48ePfrJT35y48aNo6MjFIqkYG5W7X4TibeMkU2yjjUcIupazhquR09Pz9dff01X+ZOTk0gksrKygh8dDoexqqxYbku6axprRgqXViTHq1evzs7O/vt//+8h6sVisefPnw8NDW1ubqqtOBrTpDIoEEKNUOiNsg00smH7mpiYQP74vP9v9rz/78HBwebm5t7eHmII6vGK02SRCNv7iYR+0u/WAfJOAqCNy/kGEKfVgshMgimpLW6GVeKMBNJFrIk3OEipW3S9792t1g7Bf1U4ow+Y8IiZaMVSw53AubMkLHSYQLzGxxILxBW46CbfmAkjnBDvBDCQ1+rxeAKBwOtGKlaQp7arTHUAYvPzphaVWYVo5l1g4BUKBdq6s3/gAeB1cSHTXdD5RZRSPowDW88Ouv+ykBJtNWwZb41VbAJf/Gt2Q1fZvSkrpr8yN7LZ7Pb29vT0NPgnulQejweLjDS2Qi7AKF43zdXYR+mAEwwGaXtCjTd6Xtlslu7b//l//p8jb/ff/Xf/3b/4F//ipz/96f7+/vz8PN2puBmz2bEcoHM2iHIUQv2cqWrKe7BW2Du2I7NXFgPh9/tv3rwJZ5BK8bfffrtUKvX19U1PT9fr9UQiMTAwsLm5qVb2rGesPDl0AVX0bTk7O8tmsz09Pdvb2xBZfD4f4t97e3vhcJjKTpRGkMSjGVg4HD5vh/Pyiywt9gDKhLq7u69fvz4+Ps7vt7e3u7q6AoFAPp9HJrBSqeRyOb1gYfSqp2jpzOr3MpGiBwk2FS6GdoFVA9YSrTaRX5NOoMs5b8CEjLXvqg5bYJcMWa1Wk5inkzDeJpOjn+Uhim0mySGV57Z0q9sj1y2ZIebIaLQ1StqE8Ko0CBZsxe/prj04OEglhTxlQh81J9SlVXGDtdKZtQVa+JWMxRvIrcCDdrlclL1w3bGxMfxZKhgVtbyBJvjx8bH6BeMkYol2dnaQHFlfX0diU2wxde8z/Q+qE0BK1WRWpaR4o9LKhzetnI2CPNxtNYuQsOrJy2iYKhVGgCfFbbKCLaYuKw7xWzPfa3WZkPJiMBg8ODi4efMm4S/9KzY3N/P5PCJCWBVCZObA4eEhf6IjitfrZaqgbCX2dKlUqtVqtDfJ5/NbW1t/7+/9vXq9DgKOsN2dO3dqtdrVq1e5K0UhZoHY73nWbdBG6zcmc8U0W8T+S0tL5XIZ5sbOzs6dO3eA3umsCHNOYtkCiSxFTT5JIAbqHwqFNjY2iA6kM4vMysjICHIwfr8/HA7fvn378PDw3r17n3zyycrKytnZ2cjIyMrKinYayKrDw8NTU1PlcvnRo0c0KCIM5E1LSJf2YFg3dinLa2izEnhhfMYMxkl6MCkt/Mt5hpa0hPaHKd960ZvVxsx/9/b2vF4vUV7LNokXNY5pfzB9RSw1qXLOFmKve36z4rEl1nERLdI6ZKosgXIY92340US+/CwTQ+T3fUUPmiGI4UCZVUc6M8//BqPHlD5PLfZ/I/gJ2ArZ0bQXPJHpWLCpm701EDwCn5QuMTCgqUBtaqabVFcxYkEppbqMf330batMKHRymZUYd2KGuqIUtE3nwJoYR0dHT548+W/+m/+GknfuIZ1OX79+/fj4eHh4mMrqZrMJwYE23F1dXUC7kUgEaQ1uldQo9DPwz9PT02vXrs3Ozv5P/9P/lEwm/97f+3vb29uffvrpv/7X//of/sN/+POf//z/+D/+j7Ozs5/+9KeVSsWM3UEIzpeSKScmbSopoeBnmYEefer4GOGJSfhlwbtcLuRBFhcXB18e4XCYYpNYLFapVBhrKHcSOUNUV13XAMXIC8O5AQrPZrNcbnx8HFBsenr6xYsXJDfGxsbm5+c//fTTarU6MDBw9+7deDx+/fp1ZPwQFUsmkwMDA9Fo9Pnz5yhMoQ5OLbsK6+kRc3BwgKXm8bu6uiKRCEi3SRO2looy7Ew1IjuKeszZb5YwWFV51qLCbdG1xOc1r9jGKEtrzazZVVM7BQ1SFR8YGFDhJWDU92h9nFXsF32s5c/m0ZK1ZvWoU65GBGHV2snQZDIZ4EViXp3K6lAqSru5VQuCx9G2VK0vuuc2z2tVD4HSUgCCkP+VK1cQa6YrHqsSr5MQXue8TLIRaMXj8ezXv1nXzWYzk8lAacjn84ynOGREHoyqhc5DnCDvx5+UjNEMxAkD3FDrLEHPfr8fqyKWHpHZ2bdkQWhLMJQBGE2tdvPB2R5wnEVMxBEhYBJszaygGfff+Tt/h5WVy+WgD87Pz6PyRldu+f6oYIsxoh7nxMp49Kenp5lMJpfL1Wq1+fn5J0+e/Lf/7X/7J3/yJ//b//a/nZ6e3rx5k9Tr9evXc7ncz3/+83/zb/5NOp2+efMmjbFMceZz99TS1lHCASdUZaZW+ZPToHASHqZer6PjQXr08PBwfX09mUwmEgl0rn0+39jYWCaTMb0D82zqEMw+cXR0RDtzcnpnZ2f5fF6aG8vLy+D6BwcHv/71rwcGBlKpVCQSWVtby2azT548Qd6wXq8Dmo+NjX399ddffvllV1dXOBweHx/P5/MbGxu08SWRTSkmQ4GCpVmvpaiC31gd3syDic6Wi1aRIGMTXlACkFluKluqrIYP8xun1JHpgLeUc9IPvFxVBisDgfuGNS8UCjQEsKTsLu/atycLmn/6LhC2E2po/2GRJq2MC4xa1Ry+8rrausy6DJO960zzvvEzah9FUz8YDLrdbvTObt68SZKNSgWLfdT+QbSZYbxOjk8Us8JOE1HVdFTNPr+sC+39CsgURTkngBqMiNRvhvnQAc07lMjtiYMiLFsMouLUMwCNETRkOVWmm29qq6bT6b6+vp2dnefPn6OcEY/H3W43DXxZ+L29vV6vl/pYGqcAtUOjhsYHRJzL5Uql0ubm5s7OTqPRCAaD//bf/ts//MM//OM//uPf/OY3PT09P/nJTzo6Our1eq1W83g8f+fv/J0nT5689dZbVOWQTaEw8jy0YojlMkgqjEaf6oRmVny0OSqVys2bNyGEB4NBNEuDweD4+DhsNgANn89HYrDNBAUtQWKJHsm3bt367LPPjo+Pp6am+vr60B47OjoqFouohZ2engaDQSCO3/72t41GIxwOx+PxtbW1vb29dDpdq9XK5fLGxkatVuvt7U0kEuf9F3p6EolENptdXFxUfpmUND0WLBxQxlp6BdZqlBdmBumvFaXK5srpULmKWWJqzmCLS6PzWBqBAulMK4PZUrUFgQUlo87Oe5c8TBfeKlls08zsdb1C85Nq0Oc8g1k8IuuD4gR/koyG5InNzc/pl8hPNF+xFeM7n6vl2V554IQODg5SZl2tVjOZTCKRgMlLpS5MO7N51WUOBQTndLGjc4fR/Lqac1oaI5Y3bRaySyfL7LbszC7odWiCYffxds2lJJfo5PfvjSZH6ozOL9uD9SZOYtJazFY1x8fH9KQvlUoHBwfDw8N00dRJYLUhw4QFwInmu9wkOYBKpYKIP6fyer0//OEPP/rooxs3bnz66ac+n++P//iPkUVF3paJRzMgsm74vsoonG/YKlHXwoZrKUq2c1Nqc+CK1mq1ZDJ5dnY2Pj6eTCZ9Pt/9+/dx4m7duvWLX/yit7c3Go2yNSmSNb3CSqVCmjuRSJx3inzJmHn+/Pnx8fGHH364vLxs6aUxn7a2tqCarKysFIvFUCgErRuVLzoScIB1UEezvLxMGSjCqpTeRCKR4+Pj58+fW9k8XGOrC4n+etH4NBoNU8ne6fyaZleTkowwbrUl0NHyJBZCrVyfZR/lFVperQZEGizfUcrZiuu/r8OCGq0UpQSDnF+UvTY7YbMVqcLQrNi0pM3M67aUi1EvEouWYH3sdb1sNRvDELjd7mw2i7Emvc+aJfelYbnkydX0Dw4i1xLr1Hx8E782bbFUv81ko8U2sX4vPFomG0hESLG+aGozdLdqAGL2lzAZO0rgCwaxdm7zKvqBKkfSA4eHh+VyeW1tDRRXwlhwEyFWY6zZMLSaKHIpFAq09drb25ucnHz//fevXbu2v7+/vb39wx/+8C//8i9jsVgqlTo6OlpZWaEqnTzBtWvXNjc3kShhOoGAn9+VdhXsJgPHcqVhD6ZNj8RfT09PKRy33j1p0N7e3oWFBeQBx8fHHz16tLm5GY/H+/v7/+2//bcQ6UulEjqNKqGUOHVXV1c0Gh0aGpqYmHj+/HkgEIBgf3JyEo/HNzc3STaqRXcgEKB2MxqNHh8fl8tlOCf1eh0zTaoTpIypiXzB6elpMpkEgEMmEM4JGXaxRK1sFQA0Rt/qcKGGey2dRKdrKRFeEWkpANMHRLaxIjjT9Dg5Iea214bRjClnKkMLMwfWydpsU9SuD1vFigIZLeKHGXZc5Aw6B7Al46VletxKIpnGRXxY3qMpACL6ijq6qcIbjoR5XZhbJn3N4uqY9ymCjeBU54319PSUy2WKkol8ITuR7GHiud1u8NPNzU2E9yAmm6QIK5lx0aHXAatBLXJY7wSXKk5hvZgT26rzNl96G7aCkDTyYSToMDJqrO5yuWhogIoIG0n3t5GlrkKsD7CrJuhm1kevT1xe8dlMaoNKddil8vk8St9+v395eRnclT4A4Ok7OzsU8Zk6ghRzwjrPZDKIgBeLxWQy+cknn1y5cgV9f+58dXX1vffeu3PnDhK1fr9/YGBga2vL6/VeuXIFG1ur1Wj4gGv1DTcGpJLKdwwxorckEikDMUFSZi3wGWNqLqpKpfLDH/5wamrqs88+Y9M4PDyktjAej0Oi8Pl8v/nNb4aHh/1+P/XvLFr6ECO2HYvFpqenaUNzenq6ubl5eHiI5uri4mJ3dzfNh+AzQYihjmZra2t8fPzg4GBxcRFqNkU3iEaBfefzebfbPTw8TOew4+NjqmaazSbkSqCVVCrFumV1CYVoqX38Si1T086azcD4jTobsaNYJtXqnmeVL71yWbbP1MmEYSDMEpLv2NbrjTkeFx0XGWXzMFm01u6ifABsDfPpqE9TmYYYpSLwkZo291o+oBSQ88GdcYm+LmeQOyGOxubu7u6SPSaujUQiCG5QQQYLDTsurajX5eq1HFWMHbXgTHt1s4N+Z9ZMWh0GRMRueX7nOPAt8m9se+hACUXBv8ZLBdvc3d3FpJoYujIQ/F7dgpzekvmwEv/TkuQHHORisejz+VgLor7IyoOyIhBGDIpvCvmv8fKgg4rb7X7//fdnZmbQMrJa19dqtT/8wz/8sz/7M55lY2MDwBZ3DbqLsFZu4JzoIa1qKgahKlOaSamhWRsGeIpBNwEyjZ3H46E08cMPPywUCicnJ0tLS/SCy+fzTLV6vT40NISiK5oDwWCQTCs13x6Px+12P3z4cHJyMhqNbmxsIMSRyWTm5+eTySTaYAoP6Zve29u7uLjY1dU1PDxcrVaRtvr8889PT0/ZEhmyYrEIC4WdqaOjY2RkpLu7++nTp729vRMTE/jjnZ1dz549o3OjAouWr7/lmnT+V7pXqkwxozzJVFo9Q63zvJbVu8yHlX5hG6aEzFTYsY7L2wVLYEiAw0UB8kVfb/kn/dzS13ZCdk4gyCJcQ7FXHZrI6bx6nD4SUCY5rM1OabZSvch2cBJIqFylr69vdHR0bW1tenq6VCrhzyaTSWgP1WrV5/PB1WElg59a5eaXP8TuwPRwt9A5aAyGzp/axbYcT/M3LUkHLbcu7ZpkAqVhraIztjGiDUqmT75t1Gtdmvf1DQHxpbdnQXy6eeUSnU3jCFzQ4PR6vcBNVOrBOq/VasDH+NpHR0fRaPTo6KhQKABY47yyphKJxMzMzNTUFC4vPrXJxms2m4uLiz/72c+ePXtWKpVSqdT4+PjOzg6yd0QMvBQOv99/Hv7C98CZcrlcHo8H9RaPx+P1egmNFb9jX5TGpZbEvAkac8zPzweDwUqlsre3R7+4ubk5Wl0AJJVKJZfL9fz583w+7/F4kskkwlQ+n29jY6O3t7dYLG5tbcXj8VwuR8qLZsP0lMTLiMViTNm1tbXJycnNzU2gkk8//VQ4Rl9f3/LyMvj9V199BVMllUoNDw8PDg4eHBwgUnN6evrjH/94a2ur0WjcvHlzbm5Oxf70CVM/IXEeNVE0KZX3c2b5FGJLgNtC/VA7c2oOmEd7xKBNC9Q2i0pMeeyF2+02OzFe1OrXqY/c5oZNFNgigXS8/qFJaLLoTMSmZWH6K42ImVo3myWpwM8sgVNqSzkMJzNH/HELzFEZjiiY9FHs7+8PBoN7e3s+n+/q1avZbJaViFAGbNFwOIzaRigUohxc1TpvfFD2Qtk6uxGFNrjYlP6atGjTw7CSfq97QKtg72w0GmBNuIN41hjBgYEBfIjoyyxXvV6nPFJhk4qemMMWDdxcOOKVm/1uZNyorsxms6Ojo6Ax/GZ1dZXNA1AUeWcWCzAs5HScv2AweOXKlevXrwNSNV8eVDyaz84DZrNZemlRYAg1E6iasEZchnw+f37P2GydBZ4Zrhayrfqr2l4Q92HHrQQ0pPF8Pp9Op3t6erDXNJ2Utz4+Pr65uUmvLxDSZ8+enZ2dZTIZlLSy2ey77767u7v77NkzoPNYLEYtOG9le3v7o48+Wl5eZs+/evXqixcvgFPcbncymYxEIs+fPy8UCul0ur+/f319nXocgruNjY179+6NjY199NFHQ0ND1ZfHOc/0JbN7aWmJrY/pwg4ho2yF2Lw2K21i8fMtU2XS8swdXpalpb/JVuGEI80YXC/ropIQ81ts4HC3NWvRIAaptDKWpgiD2FSvtNcEQFgBCcC39DGdX3TiPE5ioum/c2aTHqBrkYOxfqkz6JxmWRZ/QgHKei+cQV079F7AwYHRVIVhxhaEdBSIUdo+ODjo9/thIORyuYGBAYgfZ2dnkHMhRJnIMtLVsAXC4TAL5HWNJhgrtxSJRFZXV1XbjVnp7u6mGBhkTI6qE6O7JGzlnNU4swQWAlsIeUEhIpFItVrFYtZeMl78fn8gEED0WeWLqnbBOEj1yax0lU8teQDhNkJ1+NPOzk5/f3+hUMhkMr29vYVCgS4W3JXpmQHIQMl1uVxjY2PvvffeRx99RL9AtE8Vs5p+jxg1KhTSD7ibkGtNIinXOs82WpxfVOWslI5lJlC6cC4kl8uFYLbP58tkMsxaAHHs+/Hx8dOnTzc3N3G0TV+PrkLw8HK5XCQSKRaLFODAdlxaWmo2m0NDQ93d3S9evOjp6eFtUS3JtWC237t37/T0FOWUp0+f5vN5Zp66A7vd7s3Nzfv374+NjdElYH19nVpzoirwHNIgOBrm+mw5R1vCdmYO/bWE/C+a+paBvsy3TKkQiwEicQxrtStPaFFKzI3nMgbCcszNR2g/FO0ziqbutjmFhEcrj6d1eJko3qKjmYT3ix5EG5Iak0oGS/xf0rYYo87OTuLIRCLh9XqBGorFYrVapUUGx+DgYCAQiEajFOjGYjGyzdI52N3dHR4eBrBCg/S1DqhvPC/anpQf7+zsoPqLTyd5CbNh5kUG+vLpX/O/mEihExgKfo8Km1QiTl4ePp8vFAoh8qcSSnlFeBWmKgiGRX3FrNswm5GClSPph2OL/0t216TMcbbe3l6akEUikRs3bvzsZz+7efMm+wGKrOIdgaeLjGBNRdP5U5WvGc1o3Z2fwtLJNMX2THq/tU+ql492MPpE+P1+2nT19/fThSyXyw0PD+/u7i4tLa2urkajURRSoFvovt95550vvvgimUwKy56YmFhaWqJdSygUAt2mHlp9JarVKrZYwpV4zdiX5eVleiCQwCHfLUJuvV6fm5ubmpqan59no0ZyBE/wBz/4wdLSEkg3eT/LOzaHXls35ArzlZjQx/eSarOKmEz62iUZlqZVIv1gTh0xc8zFaYK8sphtLqH5Y2IUTmSzpZSgCW6YokvO5KGQVpaTbkn1KVpmVvbFOpy4qrU5tXw6/YwNhRXgdrtTqRR+BiRclj382VKp5PV6o9Hozs4O/T+3trZgIq2srKB5AH/07Oxsa2vr7OwsFovt7e0Vi8WhoSEcLhggLperXq9DmXhdzxorDEb84sULRqnZbMIYw+UHEDNt9PeVKJbDIfVHs1iB/yrBK3vteul0Q+5CEZOUr1llqmw5r4ORsXKhFx3EFnCfcdGgjVEbKP9XvjA5hj/4gz/4+3//71+9ehWqH8121eXApMkKljEbohONmVicaawtytB5us/EYbVaTKV2FelrzUgShLFQigNoGIy4XC67XC7q6Ofn571eL22uGo0Grbby+bx59w8ePKDNObQQ2lVMTk7Ozc0h0qQGGcBY3Eaz2ZSwwPHxcaVSuXbtGu2+YEybBYRmGwiW4hdffHHr1q2pqSkqiDY3N71eLy7/6enprVu3/vzP/5yGRuY8c4K/AmdlO8wJZG6Vl5zK5utwYinmX1/3cN65xS0zQ0jBsm3wa/NwZqLeYIWLIaPfmFxajYm0hNqMrVOOquWiddIcnZHyReASizMYDHo8npmZmSdPniDrQ8EBEv6NRqNarQ4ODkajUTwvWl4tLS0RXz99+pTWfHiUqCYAxFH5hgvscrmoGIhGoywB2gy+7ggTXx4dHS0uLs7Pz/OwSESwGdApRnPgtZqFthwlDqeXrU3XBAbNPR7GRcfL3DugNtKA9MAVi46zUenHMJp4AHPJWSBmwjsEHAAAg4ODqVSKtsLUIiHd5ff73W631+t1u90ejwel6P39fSrylJHW46CLws4hDMOUCCerb0LnIolpTv7OheKkpqSkSqHMjjJiyWCX+/r6Go2GnE1JAvX398diMQh/q6urlUrl+vXrtINBOZD9h01PzBi/37+7u4t4GC2o2WAfPXo0MjLC9AW08nq9kUhkcXERaKKjoyOVStGWYmJi4tGjR7R6eeedd3K53OrqKulBbY+msVY+cHV1lUID0C7EDEdHR8H7aXrQsjSLgypKYhx6Npu2W/g1Q8pfORSkm5zQljSGlkQ0y7ZaXVkttoMzw2bm6KyCHTB6sVYVE8gbUrmBTmh1lpE4l3XdluyI9jxrEzIm9qR6wiKAO71d82dziTp9eY0/DgrosCmEwFcQRicdjUof3IlYLDYwMDA8PAw2+Mknn6ytrSl0Y6ixHUx+vshBYQUTb3l5OR6P9/b2ZjKZxcXFH/3oR/A0ME/cVSgUktHELly+H7z1ImBPf/3115jseDxer9fpQYV2PDr9JlzbHv5y5ksuQgsBxCVFazp8uInQFvgkoUn/ywCCaKC3t7fRaEAsViWhoCdTpVpj1SacknelStHFxcVr167dvHnz1q1b0JrZFaSvrXp6bCC0YG4YjrxEHUjCmdLEatuCQ42mhZnEFnxPjIWkH0jON/lTJ5rTxlUxfXWrDB3JVxKp0ZcHxYeRSGR4eHh7e/vg4IA8KambaDRKIuXs7IyOMATmZBfpKEHtDLMTNdQPPvjg3r17brd7enr64OAA4MLtdt+6dau3t/fJkydQUAKBgBSUhGnysik9Ao/b29sbGxs7PT1dX1/v7e2lnTOC1zs7O5QGmEUiposhf5PkAL+H+6gJIfKMFY6ZoCoZMF6Smd83rcxFr+MNQuA2LNTXOtUr7+eV93/Jo6XzeBEM6vyh/cHHkMVQp1SkrMwyTpfLVa1WQTAwNLhXYHHwbTs7Ox8+fJhMJrFxzECXy5XJZOTowRfW9iOdme7ubkASn8+3ubn51VdfXb16VfCOqfKqehma2L0BzxqholAo9Mknnzx48GB4eHh0dJQ6r1wuV61WKWJAnAiyszn/L8M7agmUAZyi16bJb71fymFI7B8eHtJjrFAo4ClKlk9yb5TwmGhby3u7SDFY4SPuaVdX1+Li4h/8wR8gkU8oI4EBQG00sxBExZRDJWKU4D729/fTKpeHxUargQnkY5Tj2CfUbEy6sswN4gnki23fx5y+F70PSShYgjVsFIeHh4FAADYFTXqY+pubm9Vq9fj4eGJi4sWLFxQlonM9MDBQKBRQM0FOr1QqFQoFtaYPh8ONRqNcLu/s7IRCIRoFuVyu7e3tVCrFo2YymdnZ2QcPHvh8vnA4jIYsjpiEu9QpRuyOo6Oj7e3tmzdvRqNRVA0hh/CkBKEwh1oOCF4AWXW8ZnxSc7goshK6KnstfI18kbrGmPx5EwAxp90r31Sbw9l0puVnTAEAM1Q0JaXa4IDfhdTV/sas/5rare2L/tscbOEE17hXJi1P16LIjWlMYmNwcBARladPn4ZCIaoEDg4OwuEwQbSCRZ3HmatkpRBQMwmpPHjy5AmZScl0gJaiiMmsezPq3tnZ2dDQEMUHQ0NDKBMhCOXz+RqNxtraWiKR0F5ymXNe5o0TpCpWaH9m3mO5XFYNS89Ley3bp84DLcWzzIOCDLILZomZUAFOUiwW4RzLjRNNUCxvMnPql01hIBszKUclseTU8shMVBxzelGxZX766aeSbuZs/Ixh2djYoOv8+R4gIZvLTHGLbmlW2SpkFsWdRDbZ6lwuNz4+TkFKKBRCFBxMo1AoAHEEg8FcLheNRjOZjFRBstks9KatrS3KC9fX14+Pj2dnZxcXF+fm5ra3t//gD/7gypUrX3zxxVdffUXsBhEVUELyJmZSxUQMoEbhR3/44YcU8iLCQo8r+t20HBBMrfQ0eIvW9FWyDijGGk/si6AuNcfRBzTmskQmmbTj+zhMBmH7bNUbOPJvdrTZTr6Lk24mIZ29fftfHjA+c7mcs1uK6rCpj8C/TiaT1WqVYhbKKMrl8uLiIi26qSc2Bw21XpMgqNZTcAnwReB+oRkE7imVZ1wHwZUk315rHKQCRDA+MTHR0dGxvr4uiJKSPFUGwd4Ta6v9ma3fmM+uFDHAhfNUTD8xIHlAZctOX8agyl1jstWV0OLFOhPakvqz0tfqron8dGdnZ6FQQKAC6j0wiHBnCi+0YXDPMNwIqoSWEJmxp9L9HVY7CGEsFvtf/9f/FYNG6gXrQdAfCoXQEuGhzgVbFBSYbSYYSrT+NI5MCNC3er0uHQz1lub50azC/p6envp8PnIpjUYjlUqhWYolJQmJ651MJm/durW9vU3vgqWlJaYIk1V1REQZyWQyGo2C3x8dHT1+/HhwcJB6AVoynpycrK+v07igWq3CplIcRCsH2Dmk5ru7uxcXFzs7OynGCQQCiBXUajVQkf39/ampKXSdzIPxUQMeYiWxXIipJaAut8KcQNqBrZrylsvAjEbNBKAFODiV56y0JLgzzH/tAfxL/oAPAJ+Z2GKbKPiVh7NC5KJPWqCcKethPotCEFOe0KRzWRdSkY6oikpj+P1+UMJyuczeDDCKt0tIC/m3Uqkw+dmDDw4OqCEQhQDwEZtLWkwGCCjPAlLFCwQw5G3u7u7iQ2xtbb3//vtdXV3VajWRSODikEaDjSCU77UO05ZhICQMTZBer9e9Xi+ULbwW83XoZ+cLNScwtsKq3NPPgNHEnc6GROAMpvJ790s7i9GE0au4ysyutYz5iE70gEriiaONFCjKE4C0lCPJf4e/6PP5cHst2QmSjR0dHXShZEXjbouJBBSmZ3S73dvb2/id4GnAvIDJPI7X6/1X/+pfITX68ccff/MA7BLgKaZeu/nATo180xbwSxC9WCxG5brL5Uqn07Rhf/r0KZendNDn81UqlXA4DL06k8mUy+VwOAxLiT3tnXfeWV1dRb2agmwYHT09PcVicXNzs6+vLxAI0OGFVuVHR0fPnj1DOLtWq1GVozSgOY3UrQaMhR7YFIYSaYLMIKDO1mJuV9YcNfE4C+iw6tks1Ls949jKwltaSOY9tDegLcEB8x5EQZMortnG7G/yEKooi6YAzkQ53+DQxubklbMhsXtRckb8Cz2AvgSlUknDTpUv2j0PHjy4cuVKJpPp7++newhbr9WaBEaHmWG23gLzX3I/OprNJtdSC3CsjLQD32w0NCb4PXBUsGsyT5gk7FQb8rt5ON9OG19bYL35e0yQ8yr9L3s8gg+YMJ1plDRnrLyi+KlmxYNMNkez2SyXy9VqFfYeRcVO1ReVgFuyM/oM1SR4Boyhyk3Mz1CY+l/8F/8Fdokzq33l3t4embxQKOT1ev/RP/pHP/vZz74ZF5MNA1guGRcTvDfZsi2PRqORz+fhJ9LTNpFI9Pb23rlzh782m83p6WmPx7OwsEDXeijovADAh2QyWSgUqO8cHBwcGxt7+vSp8pZer7ezs5MciFolLC4uHh4ejo2Nqa9SIpHI5XLooYjurvvUbjQyMoIW9srKysjICCcEJoM8y/6P/Eh7MwHvnYVkafbjB2l+dFz6ELlY5X/WBy6/Vs3UgvWDuFzExQy1YBmzjvEyF2oZHFjiKhedSpaa/yr4lYW1FomT3fFKJF1UKnPThZjU19d3eHg4NDS0t7eHSoxpOnGZhTz09PQgvXRwcPDgwYNYLDY5OVmr1YrFIpPf7FhGcQDG2gTKrJL3crkcCAQqlYp5z/V6HQScIhGcDElofhdjrXw4Bd+4nypFuSh/e/lDvQhM8VKl3PUZdfgTG5idyXqPu0Zeh5SYTmjx4k0pV/5FPxZbr9Quw0jGi8lAggqhJSJ782UxZ1RLYUa0BP0couJxz5StmuwXLoSvTX8rQrHDw8OdnZ3j42Ov14t9L788/qv/6r+6evXq+Rhqxqv7J5s239fKaUOMNwsgSfIWCgWopp2dnYuLi4SKRHnNZvOzzz5DbHpwcPD69esrKyt0ledyPp8PNv7g4ODCwkIgEKhWq+VymSw5GR6mtdfrLZVKgEq8wvX19ZOTkytXruzv70NTJ4izuvriOYJf37hxI5fL4fVPTk5SbKoeaNFolFvde3kQpZq9/pxTk/VszZU30+//6ztaLjz1+Tapct9R/+F178oSF5SZNqkF1qu8/BZialpaJp5OF+SUisUimCP8KnYvnBjy0ipZxL+GBXR2dvb48ePj42N0b6yMNKpA/GAJsFlHpVKx7DWoFAaOeS5RwO8oDKIcO/RW8X9BV0yBgTc7tArIxZlv1vqM+YJUYGlqgHR8S1UEXrBY82ZqxzRTuqiShOLPmc41B41HKOOgWNTskKUTMi3BoOS/mvuKmfhRN15LKwIDiywMrAp1NYMO2Gw2V1ZWXrx48Z/9Z/8Zolrn+TC+rPJ8JEWkaGy6MGa9A0R0cbSJyOCypNPps7OzQqFACczJyUlqKJXJZADsE4lEtVol5bq8vDwwMCDIqdlsoiR78+bNJ0+eHB0dRSIRr9eL14xju729PTk5KfXU1dVVVoXX68UmVioVtAoRxHn06JEE2tnDGWjmqMfjicfjT58+7e/v9/l8S0tLiP+xgdPrAWScVcq1SP6QTjQxONwcU/vbos1YVUnt3cyWrqiVbDS13y7iq1msbSd4baab6vW6FL+4eaIZk6EozMQpLGVCNIheAtQ6ua7WbVgJN/OcFtpm/UmMgpYtLEzXlV3WDB0kGAKOjIIaSmH4PpwQZ1OdT5XmwlhzcgQr4OQxSZjSeha/30+SiviXqzPOuknVcezs7Ljdbo0tzrvX65XgpUgg7SkQbQ5ztrBDswPR6rO3t5cb1jNawJ0zSmt/LZXUKcXlrDDS5AE9l/S+lEKbLzNPciVNiMksRDC3eS6k7U3EOJlpXpMwq66urq2tLcBAiJsQ8swyMWeQZy1SQh8xgy3ql5KNeK7Yru7ubpgn+XyenoKPHz9uNpvvvfceBmd/f9/v9/9uZ26JaZrccnP7khat2ceMdmQbGxtMQQoIo9Ho3S/vQvzY39+nP8DAwABQ+tnZ2fz8PH4Ebv/Y2Ngf/MEfZLNZHqxer29ubpZKJbfbnUgkBgYGMpnMyspKMpl8++23p6enc7kcoYTH41lcXASqrtfrW1tbyWRyenp6dXWVYcUHl5ObSqU++eST7e3tTCZDhyT0RsbGxorF4tnZWbVaZd8yMyQsY/k4ZmUdCVU2A35jSpSZSe3XXVct18lFH25zkjYMa6eoqdMydrzRYa2c1/qi6S60BEOleAcG1TLyU/sowFAFRni7FDKEQqHl5WVCt0vengoluru7t7e3UfYwNQvNGyZiA8SIxWIH3x4XndyUY7X0tS111u/o/FK0zVPg0wlz4xEAG62vdPw1HBJsUPtH3GERFk9ejgOuw0UhhcnjvMwNW7YeF40tCto1HrTzPBbMYJkImJ0mi9c6sJakQ9i/mR6ska+++mp3d/cHP/hBMplUq7Nzqdg226PZoM+6kpiDyq52d3f7fL5gMHh6elqpVCgnaTabDx48SKVSIyMjc3NzkFFQ6uBhxsfH+/v7t7a23G53PB7PZrP9/f0gd8VicWFhgRGZnZ2dmZkZGhqCSYLczP/1f/1ffD0UCiHFFwwGyRZyrY2NjaGhodnZ2bm5OZLyzAA0rzOZDDJP3Mbm5ubp6Snc+8PDQ0SgFJWYj68iBQt9lgdKgGIi1NrS3gAP0bJpT/u/5Kms2Wn+CQqEGgS/MW5jUQ+VI7VUBp3falmceZHqLD/4/X4EuGkoIa/fzPqKA6C21lb9OuBDKBRqNpvOnFLLQ6VkOMugqCRIrHFQ9aMkk3BNTN//Inl+DjXM/O6gh3UwXOC/XAWvFmSWvRDqy0U56jc4lIG3UBFzV+ZOlMDgrZ0YnOX2fknLOOwyx+DgYKVS4R1htaGdtO8aivNqknzUmv0iB6XZbLrdbnYjEgZQXEql0meffeb1ej/88EMsNaAF49Du9auBscnZEudBn5GcE0XbACAffvjhL3/5y93d3ZmZGcxlNBql+AVqJ/dRq9VGRkbK5XI0GnW73ZTz/umf/mk4HN7c3Ozq6orH4//gH/yDx48f5/P5WCw2MjJyfHzs8/nW19dpyEvNYSQSqdVqH3zwwWeffUY1MAHd8vJyIBCgYzGVRXDAuf87d+709PRMTEzk83kaeg0ODpJbODg4gPNnlRTKEAAkmTZFjZEukm1quSZfaXlbeotvcFhiT9Y0EuHfqgSxNEleNzIwWQSXfIrLlFZqx5L6Oympi86pCkBRElWAAEbJgiF3Yjq8FxVPC0FSeXQ8Ht/Y2DC1rc1qZrOfujCHlnIl2tetJiwmfvo9HsyK3d1dUqasTVMJ2uoz9zdzmDUH4Aadv+9it181baA/yyk2d6DDw0MoavBGNjc30d6yFM0sKNLyIaAjy75dtFiq1SqPpmc5PT1dWFh4/PhxLBa7evWqz+cbGBhg5qBiff6CWH40kBVwJhwZnRHEs/f396nsAuJRgbywPLYmEnH/5t/8G7fbnU6nycNSnLK2tkb2r1KpwGXJ5/O9vb1ut7tcLpdKpT/6oz+an59HTmx2drazs/Pjjz9eWFjo6OjI5XK3b9/e3Nx89uzZ6Oioz+crlUoTExMej2dtba1arYbD4X//7/89qtZwPAqFgsvlKhaLpVKJYhkgRajpU1NTjUZjZWXl+fPnY2Nj4+Pja2tr/f39IyMjJBiB74mGWqKoziahLaEGi8lnfth82frZCmVMwkBLqPeioyV1RJsK/zXDNxwEEWy03zirKF95UY/HozaeZubHumcLbW/vMTmBEVHacQbNT4r5zhs3iwlMOQ6GF2aR2jOa57nowQVBErMPDAzQmpnD3O3UGcv6jdSLnArd1kV5OjFHxXdSAPfGxlS0dCXcSGNCSOC9b21twWW0nv11D0vgTIdl72Bu0PoDrpt6LPS+tH1qKKPcILeNQSdBhcMhATxrgjmvzjBSFMO8ZfMGNzMvZPo6Ts6JpqjSAM5yZU6Flmc2m0VR5PT09De/+c3a2tr4+HgikQgEArQEYn+iyuacZcdcV2ZZVerocphyP4ODg+o2JtfMJOfXajW4iqjV3Lx5c21tbWVlpV6vw2B1uVyrq6tTU1M+n+/58+cEsIVCQfpV//pf/+tIJHLr1q29vT0Kex48eNDb23v79u0vvvjil7/8JX7r6upqMpk8OTnB+e3p6Ukmk6iUcZPoqdZqtVKpxAZD08zu7u54PD4/P69OBXyXhvAffPDBysoKAA68FDwgi/ZnSqOYc9HZc+iV9vQy8/4yhvK1qHXKczo5ZM5bsq5uUg8vcvSYMKYG/GstbyfOYyrBOp8RZDAYDJoEDBLXXBc1HKR/iE/bl1A7/dyLfq+TtEG6nX1k2l/ReVATgGkmLhTXu1gsUqnwBsgYYD0rCARPYiDITF4evv9+D8y0ugGYmUnXS3kWdYWVHyMyuCndp/ZMzulkIc7OtdPV1ZXL5W7cuAGHxyzY1klasgaVhDc54JYph+FzfHxM74hms/mb3/wG/f1EIgGxmqQo2zPIzLk2COdSlbP4oRRr4VOrf2VLui5MNY/Hc+XKlVQq9eLFi2w2m06nIZZT85NOp0k8gmNA8kdFZGdnp9lsejyeTCbj9/vT6fTm5mYkEoECeHZ2NjU1de/ePQK0ZDLZ1dW1srKyvb1NnhRl92fPng0MDNTrdUbnyZMnNOOg6QYauDzC+vo61BHAOBRVdnZ2iKZHRkbYhKCCKGLVS5KEowge2swsvXCTt9+yjNB8i9ZEMdOSFyEAb3yo7EXG2slN/i6XYECQizMrbszPXGS+X/e6UHqI88zWkfjOZOQFQYCEStDcec9vcOArkdZvmcn47ofKicl1SxdM7bQRc3+tQ1Gm3++vVqvyGYE96f73xk70JY/275rUgrmmdl8aOLpsY5rVjkD0OzMqVW+2Nli/02pzoN0GLwgSp9kmRl9UTaw5VlZ7P33ejI+hq21tbXV1dd2/f//w8HB6ehpRaGwjsSDvt1wucw/nOXQxUczyPGINGkcyLZiRWECTB4rwDRWxzWazt7f36tWrq6ursVgMPWu/37+5uUkycGNjg1SMx+OpVCrVanVra4tKgZ6enq2trc3NzWvXrg0NDb377ru//e1vq9UqAr6hUCiZTD59+hRou16vB4NB+poDZx8cHIyMjKDtMDk5ubi4iHtFB/RgMIiLHQ6Hs9lsqVTa399Pp9NDQ0PZbDaVSoXDYZExm80mTD72MB5Kg67CJ8JYq2j4DY6Lwv+W9tpUQ72IsdceSTCb25tcN2c1eUuC0GWeiC3/Mp9sc34nn9q556EzQ1bQrLpW6xb8btMtUsNJM9l7UZB+0VM4s0z88F0s9UV7GzU4AtOFYvl8PuToksnk616Ik1CFID0KMCIcIPgJuoG/ySoB3DjsjKlV2/3S7IJMmul6fG3p/+g8Zn7yoms5adrKw+VyOdTZzFlhFp1Yr8ycEsxV85b0V24bgf5qtfro0aNQKDQ5OUlxPz41qi/QBwuFQqVSQaHlm6IYWHcAGgCXe3t78XicChzQE0Iw+ICmHZHYLjsSXbqPj4/n5uY+/vjjjY2NnZ2dQCCwtLQE/jIxMTE6Ovrs2bOFhQW+pVw5RbdPnz71eDyDg4Ozs7OfffZZqVQaGRnZ2tpioxMYhNO0t7fXbDYzmczZ2VkgEMjn836/v1QqTU5ObmxsUJSYy+WArZHuc7lcb731ltfrffr0KfqoVFFubm7S2oaiXuB/rICZXoODLLvGn8yN12RSO3OMzi52F8XdzmavstTgTqalVo249XXzQpo3HOiA6wy6N7pEiyJqdp8QXOjk3uqTLXcXa5GYA3WRV2vuYS3ZIFSL6ZFNqEqqishsOhN35r8tr9j+fnQbpipOG4vWXgisDY9NQxoMBkn9sdmoDWYikWAhmPd5mQAF91k9uYXjVyoVyomF5LIQvkd7bdY6WRwnnB6zzRAiTaffYnemVySaijxfM0diKT+bG7P8LTOhgqQEiwhdQ2qjQK6l32KWyZgdAyKRCDA3chrqkWI2JOE+qV2s1WobGxvXr19PJBI0NCBxCMm4WCzSVwgFUAKL34l5SkTVxFkAyKQawbSw3pzZ7aZcLpMHgClx586dmZkZmk5++OGHT5484SSff/45tDydhGvxm0aj8Vd/9VfT09O7u7vXr1+HoI1iamdnZzwep+YqHo+XSqXDw8OPP/6YNhzw8EgJIrtTLBax5iMjI9pp2WyXl5d7enomJycbjQYTNBwO9/X1oaa2trZmKo2ZPrXVWUqP8Fq9tZwZm4tKYyyLrDltVoGL8H6R9sVFVZRmR04l3BTHmcl3swznlV7z67YZa48Utfyk09xfRFvWvDIt4EWusXVdy12yfi8XT7/8LvCRadAtSVUKgyU2RAqXAJd5rqY5ZnGmdT/W/bO7kzcyfy/Gekva7v9PDpNy2mnUdqgnqhhrMuXiPsg1aal1YxprrLNaEDSbzUajAdlX/QTMXCsIKpeGPNZoNFT1DS6HY87HTMcCVsXY2FgoFELnw+12S0Jd1hlLrb35Gz1rkxRsTlB8WLUgo5OhvAmtc6AuOtsCSty8efPFixder/fdd9/93//3/71YLIJmJJPJ7u6e58+fm7V/iln0XxKV6XS6UqlEIhG1YMD3F2ErEAicnJzs7OyYIXBnZ+dbb701Pz8P30VdRXAiyC+hoy3tx1gsBiBYLpf39vZgi8OwJrAwu3ZdcvrKmrT8vLWKrPDtIl9MHq45OyUQYQrVWzid3mzLjUFT3Izd5JiYtVGmBX9diojTg3aatlf2GRGa+WZH+4j+zUztX1ORiA5iHSqwVNkoFoH0p3CWJZ1BEXybDQ8xBnp46/eHh4cIDZoyyNoqrKRZy6M9Z9y6hzcbjY5W5zE9dNk0nEvJXsvJdUaEbPYoduE+VyqVP/uzP9vd3UXpW6lOM6sM3kDtHo0xCbuhwNGYgjcCw1j3Sa0JXdk8Ho+MOwgBEvwIh9Erhyt+o5yiBsYAmnoGk5UNUUnkQY0RoyDKKjLW5FK//PLL+/fvI20OQdDj8fz2t79Fb4/2jKZvaBLjCdKDweDjx4+xnnt7ex6Ph29RN8zI7uzsUJ04OTkJGE0tIrfk9Xqp7ESmFZHcRqNBDU4gEIDfghePhaJ3Mnbc7AjeRlHMnARW5HVRmfX3RaA2BWUEVphaOa8kpXAS/DKzlNasKedFc89S8e74/g4nje8y+5z5AZOGaHrZJjzyxop9f/OHYjh8FGnRSb+JOi/rqfkNoiVtxhAsC//OTLfidUHDaMma+L4OUzbkoly6eXT+/l/1dSsRap6KxKypBWbqPTgPFE1RxVheXo5EIjCMIQojfQH3mYP+J7RnBN+g03E4HAZignmBAAgiAea1stmstJ339/dxEzVvsVdsnPRpOTfWQLQCYhQOaymydJHrhUdCiaT+SkQwPDzc29uby+WCwWCpVBoYGLh37x5oQyQS8fv9Ozs7gDX0GTg+Pg4Gg4rNmZHKpeLOw19BHzaRSDAKy8vL4HR+vz+TyajLEcliLCzb0VtvvfXixQt01CTyNzw8vLKyArAeDAa3trbkkJ6enrrdbtgpGxsbWhvU1jNrLS6ONXX4rzr0WEG3CWI4PXRFMKqGAi4ExmIVtaSRmGAI4lnmAuDN4nmxLak1hsltAv1AJYOsjjpkKp9p4vJyx6wsuehTl1/h0pUNhUKUPpugv/lJfm8aYivDY+6pbTIBOpu2tIuM+EWe/mXYI69r41o2hySjQ62Wz+fLZDL0ksZNq9Vq8XgcpxtaFEMUCATUZbTNvTECWvvodwtaBflEHxiH0Ymzm+nQiyAXS9dMQ12v1+mJJYUmDkWKJmzV8fvYGvobAugqlQqUGNNqKQpRaGi2SXHeJ1Mamwt+nU6n33777b//9/++2+1uNps0k6LxI3pbqHZQkCHAqq+vr1gsErgzJ88rxV865pJgoviW91UoFEgEYsFMTStWPQN4HkNzo6pf0LjjQYPjqCSf7RqSsvmQiURiaGjo6OhobGxsd3e3WCyKvVcsFtH9Q1lpbW3t9u3b9XodYMT0TCUiiuGAB+J2u4F+PB5PoVDI5/N8ZXZ2Nh6Pf/7554lEIpvNjoyM1Ot1rg7VGm4fc+K9996jrHFpaWlvb48O0319fZubmx0dHbFYDMnwYrHIDobzLhqTDIcJ+7ZPB5ufacmcd36lZWrOeUJnHY1Zia4Ugs5pbSqyrQplNIPNhA+gkGRylcCU2BD//V6cayj8Xq9XLVQu8uacQL81tibrw/ywuYZbnu3/g0fL/SOTyQwPD6u+gcJd6idZQWdnZ5ZORctDlBiSYOzTXBQhJ2263wvIY/YnkSUySzQ1/527o+WUdHzLEMf2gQrg0DhfrrMIq43/ruYhpARo8P3ee++BONHmOBKJELV4vd6DgwMqUciNBQIBfESwVn4+OjoibVir1XBP4VCQumTGZjIZgA0CYlVBWyP2O2OtDKHpA4L14JRhSQFDLY1HyIChUAhV6N7e3ng87vV6gdsDgYDX6+3o6JiZmQmHwwsLCxhcXGbTDjLbcOtcLhdb1unp6Y9//OOnT5+SP/V4PH6/f3x8fHl52e/3X7t27cmTJzCpy+XyzMzMD37wg08//RSlkWfPntFv4ssvv4QKQwVBqVQSZYXmvORwoVcPDw8TjziNtdOImDOppWjcRYc1aYQUW4qR5oVMtNqcgiqEs+alsy7u8PBQGQ8lWi0uoHk5ylbN31vJKyKSNzDZ5q1SwUSToFdWVFuJL9NkO8tEncdFCcP/jxyWoJ3zWQ4ODu7duxePxwuFQrlcnpiYQGaHgsPLdNfUIeH/iYmJXC6XzWZBsVWXi0FBz8Rim5gnuWR9lnljvGU2FaJGtXCyvii71PH7p1IJDMk96p5eyedxHm1CwN3dXdq9QpKhVSwmW6rLdONkGJHgJxxB+oMqSl4KgMTW1pYW+/7+frVahUYszSLwE1NQyOwc9jtj7RwmlWaBo8ueiuDCJxuNBoEDYjrEEYgw1Ot1n8/n9/tpUby4uBiNRpeXl1UeqsFln9d1XS5X+OVxdnb2/PnzZDKpVO/+/v69e/cikciLFy/q9XoqlaKzTDQaXVpagqjIhBsaGlpaWoIymUql0NCC7o4WeDQahbbp9/vz+TwuHh2eYrEYxhro/K8Jv7P6AFhTX9a/JYvD9KxNG2rNV1NDiuVnXYLPW91LyZlAZDTrcc3vtmeGXH7rAhl0+oMXcf4kkdPy/G1I4t/j8UqM5a/pECTSbDaLxWI0GqXd7c7ODtbkki2+cABxDrq7u2n1u7i4WCqV2NRNcPZ7OZwGVICy2XqcJinmfJbxchmtXiS4L+Tkeykc09rh/PPz80dHRzdv3kROIB6PV6tVQvZvZPB6elCOI1VATTV+ISFLKBTK5XIQxumZhzAIqDI9yxXT4KfrTswC4G9yUVCkVepjCReodwzg6cnJCa/T9IC4A4T8Ozo6NjY21I8cndZcLsfHYrHY3bt3AdSsNldinnBOWoJRRgxDfm1tjbwqVfPXr19fWlriDLFYLJVKbW5ubm1tVSqVqakpCtyfP38+Ojpaq9VoVQkHnEZq2l1CoZAan3s8nmg0ur29jTkYHh5eWFhQPSd7Kc6mKQxrzg8L3BBho2Vsjl+vliUtM9ROoNxZKqIlJwxBrTz1SWT1WZ/m7MSpAamEQs7vXS4Xr1XJZPhDJnCp7JC15ViFfM4w1rlCSGdpA1AnF3ElLV+h5XnMUtKLtLZNPoMQnteq1tN01bswxVva+5hW5Gpivi23ZIYXB4j8/F/8xV+Ew+G1tTW/3z8yMvI//8//85UrV370ox/RlhdNhWfPnjWbzUgk0tXVlUqlLPkXdjsKSfhluVze2NgQsYE7Ucco04JcNObm8zp/do6/yUblB+WrQBX4r1U+fmJkR6y+S+giOcffJLGYhWCiNjnHfHd3F8XNZrO5s7PT3d2NeP34+DheBQQzAW5mLyEuhO42SxvZfelzFAoFvWjIZs5IXaZZXC/JnZ+vCmpYWSGmit4lvQbWMI6z2IhIopAGSSQSNFEMh8N0v3WexGwaiwIA0vWbm5v9/f2rq6tgOslk8vnz5x6P56uvviIDsLCwsL+/v76+Tk1jJBJhezg+Po7H4yx4Lur1emnU1NfXFwqFSAMS5VF93tnZCXUEWgucSuZKvV5n86CtA8AOv+x400OcjZbcT61VEenNz1xkIKzPCO0yezma6SCZeJFS+e7JyUk0GpWMnJOzZbYj0M3ozOIIvVbxGx6NlMHpYADsRnEBCZz2Uht/A3wPk/Devo2ZdbQsgHqlTtbAwMDo6Ojw8DCd1FdWVn7605/u7+/Pz8/HYrHe3t7/8B/+g9frnZmZoYqYrEy9Xr969SqWem9vj9QxVyyXy2YFcjabZb/M5XLmvvvX3X5TW7u8ZnJj0F2sypfT12R8C23nBxOCN4+LUM3e3l6o1lCEu7u7KcobHBxkYqu00tqMdVphxWZaToUsTl6K6f8BDIh5oSbm3/S3Z9HqYtTCtJlDWvnSQcaNPzo6Gh8fz+VyCA7A9yB7yxCY6uzmYbJBaGLr8/loKN5oNKhdPDw8XF5eRv1ke3vb7/eDE01OTtIDrKurK5vNvvXWW4VCgUYz6+vrhULB7/eDd5Mj7e3tffbsGUV6NIupVCokRff39z0eD41sqOvhAfkW2xJbq2aYxtqZG2wTi4kf7XQVLSpeeyj2MhGfldsUVG1mwC3whP2V+IbfMMlMTq58FrMexxqB1z2UWuGWdnZ2aLgVCoUYKPXW4jD72pnD+NeNR1uMiJYxPnO+JaFevry2arMiSR+WJ0hhYblcPjs7CwaD6+vrV65c+Qf/4B/gLtDMaHt7e39//+bNm+FweGdn586dO0+fPp2cnOzr6wsGg2dnZ+jq4NPRQIMLHR4ejo+Pt+mEYGXqLqKivtahLhDmShEllODMNAjHr1KVsqpDLL6T6dpbzrUe0Hyi/v5+SkZw5mC4IZgO7brlvqvGQ0r2mgOoKg0zhnaOqqYE3V/NaKwHMV+ncdETWo6eIn1ZE1LJOM57e3tTU1NutxtlRY/HMzExQWL0xYsXvb29169ff/z4sXOsu7u7SfSh/zc0NDQ1NZXL5ba2tuLxeCQSyWQyjUYjkUiw54ONjI+PP3v2bGpqKpvNUjhweHi4sLBAuWcwGKzX69VqlUzm7u5uLBbD2jYaDbrnoRtLHvLk5IT2COg6gW4zWNwe0CqNrsmMqbSnjdG8aEJrWcpxdjI9TCPrrKBrqY5kftGM2a0UueaxxB5NA6SG9yKEOHdu2REhDx3f7cAcS2weEV0qmOjFJc+6JV3M7HFlMlu+r8McXtXOtWH+OT0vsxKvJQe05eH3+w8ODra3t3d2dm7evPmTn/wED5RW3OTMr1y5Mjk5SW+mO3fuZLPZ3d1dGK6pVOoXv/jFzs7OjRs3+vv7x8bGrA5HZOMLhQIFINbzvtah9MlFkZ8+IJiOq0gHFVqwpoG+2GX0q7LOadlEs+eyVbqhUNJZeWDBlXTeAifY3t72eDxiuKIeehEcp/JpgBrgCv3JXH0WRdVEhk31UxOO7/7kk0+UTBBcInIPd9bb26s4hWZ0qp3hdL29vdPT0+T0l5eXR0ZGqJukq0uhUCgWi8fHx8PDw8h3gBGfnJygbCfGK88fDodnZmb6+vqy2ezY2Fgul9ve3sZGJxIJlm4qlfJ4PHTILZfLxCykrQV7rays1Go16l9OT083NzcnJiYKhUKj0RgcHPT5fOVyeXl5uVAoHB8fv/fee9VqlRRNIpHweDxbW1tU6Cs2kWYuOwqjqXETwGplz82CTzP+kvyblR6xBA20h1tnMCeBOflMNrfogEgakTL9vd4ThmKvyi5SqVQ2myU2knCu2Xxaz6U5YE59r9f7BilZQYoMuPpkaucgeLQeVgCryarkNVmr0fyNVS4rJ8v5LeehiEQ5Md289V0RhyXQyBrWSlYsbAHZFuvm/fff9/l8uVzut7/97ZUrV9SoF0o+fklfX9/Ozk4oFOru7i6Xy52dnRMTE3g26KaNjY0NDg5Wq9WDg4NkMhmLxcjj4WNBZl1bWyMfU6lUfD7f4eFhLpf7+uuvzTSXk/mu+WyuBdMMOV+EDF9LuJVYwZzhphZNZytfxDw5Vl6etVwNU8XF9HC1WHgpbrd7ZGQEJhuNYVkmxOXIURCXSI3PPKjE0WxkjTsnmDwkKYmbULD4F2RQ+/r6KBLsyefzGEo9m5kINncPtfXVrXAHu7u7vP579+55vV4o3BAMWfaFQoHm6r/85S/ffvvtsbGxbDZbLBbFdEHro7+/Hx5iLBbb2toi9OO9gp8cHx/TKO/w8DASiQQCgRcvXsTjcaj7gNRyexlWsBTSaD6fb3Fx0eVypdPpra0tHGfE9gYHB+/cuROLxYrF4ltvvUUrXtOIXMQ6MN833G1G73XdzItAp+/LN4Q2bpL/VACl7K4oz9DCmHYXsTIsXvNfByPi8mP4NwOAmN6QhTZ+F3j3lY8JUw0g7uzsLJfLLSwsUHwYCASi0SilZ2dnZ3fv3n3vvffS6bTX67179+7ExMTY2BiabdCxMpnMw4cPT09P33rrLTisOzs7ExMTVDZjJU2yzevSAV+ZdDXDdPH21B1QdeGysCYp++z3Kwbkh1oDaOpft3RdNXWdQSdVGshmAZ37fL6NjQ0yVeFwWKXh9EJrWXaEHUA5D7ukhL9zdzFJTdpa5ArzFfh/574znA0+x7/kGzmXJSMpv8CcssQL+/v7AwMDqVTqyZMnJD1nZma+/vrrVCo1NDSErN2VK1eePHkyMzPjdrvD4fDp6WmpVNKYxmIxBECo/6FEhSR1OByGiZHNZsE09vb25ufnMbUzMzO7u7u0E1Nz0vHxca/Xu7q6Ct5EUc/p6Wkikdjb24vFYui1UpuEOCFffPjwYSqVsrxdhY1WEtmM1MjhIL51EYD7/fLJLmmYFMfgdZptU5hGolHiX9B9Tu5Py+ta+R9zuF6J17/uoTRpG9UUHW+mxH+ZQ41arN+bDbdMl/97SdAx1HC/vF7vs2fPksnkjRs3fvSjH0UikbW1NchOgUAAW/ynf/qnqVTqxo0bV69e9b88KDgSh6q7u3tubk7SfWKVuVyuYDCIeJDZ9PmSg2kyLqxfOj9mAgJqnmIqeJgwrJnBPvt9oTsn+GalEE1F05aooFMQeHd3N5fLMeysgkQiUavVtFiw1xjrlkPh9XoBG0i3VCoVtUh3Hua2IQxHrdbRjIQBcS6tQYGNohK4KVq9ZjtBaQTL5+JAZADdaqSfQXyAn7a3t5EBwSweHR3Rl4vI65sd4yUvolqtzs7OPn/+3O/353K5RqMxMjKytrYG6/Pk5ISmAQzZ8vKy+v7GYrHl5eXztjcvdULwbZGzOTs7y+fzLpcrFAqdnp7CsymVSvAop6en5+bmKM/p7e2lQTUEyUwmo6ez4H/nRJSjGovFKDy15FM6vvNhJiWcrCNrxzbnojJalmKZ8jZmw2zYo1JbJevdkvHSMufTsqmg/tTeeH0v29hfa17xO0YPb4zp07Ijn897vd5qterxeH75y1/+4he/iEajdI5WBxzUyjo6Om7cuPFf/9f/dV9f3xdffLG5uVkul6n49fv9V65c2d7ezuVy0WiUIrqtra3x8XFyMMiiWWb3e9eztlAjE6loSexj5I8MwlxLn9r6lplaM2e4PmaZadhHR0dHi4uLYERYPGijR0dH5XIZr58eUk4HhUPlMM1mMx6PizjbsjjZynlQTeN2u3Hwu7u7MWK7u7vnRhzFcYjZ2HVySlbZBWQPKT2Zo+N2u6PRKCWV9XqdAG1gYICUIE16EKR+8eKFx+MBqchkMvA6IMlRel6r1cLhcLPZDAQCh4eHJLsbjcbk5OT+/v7q6iq7XE9PzzvvvPPgwQM6dT18+HB9ff3WrVvQ8mZnZ7e2tubn5/v6+mKxGEqSbrebDWN/f79YLMLDAx7BfC8sLIyNjYExTU5OfvHFF5D5QItEVrXgRWsB53I5p4y9CZjoJGZQZs7IlgkT09SaGsqi5ZkqwKYwiHlv9MoxtxCzNIk7kQKJujqRAxDIa+rvWMvPTOwohmVCO7MxVsLTHBmNBmfgAU2YyHQUBKlLq4wyP/YSkJ9AIGBeUfLN4uQI2XduJ/yJodMcEM6oz5v5RpMfaX5eD3jRjnKRDsnm5mZPTw/ywtB+CYCeP38OkCj5aRxkn893cHBw586dX/ziF7u7u/Rcp1vTP/7H/3hmZuZP/uRPcrncvXv3IIogTxEIBJLJJDI44rGYkGvLezafxdkTw/yWOb3FLBKyhAqQPiB2r2WUzwwvxFRWMF11+ae6lpUEYpI7SVYg+Gr3LnceLLurq2tvb29tbW1jYyMYDEpOGW6YiAbk+Q5fHru7u++//34qlTK5N9a4OZ/o8PCw8PIwV+s3xuqdd96xEgiiZMgjkzA5BG9rrZ6enkajUTjIp6en9BDjVJFIRCU9YMqIxfj9fog7MOporIV0t8vl2tjYyOVygUAgm80ODw9vbm5CxsAdHh8fHxgYANlgb+DVsgek02m2n2vXrvX19ZFgHB4ezmazQ0NDQEjJZBIGCKOcTCZx7XHeS6XS+vr69vY2KLmSb9qKTRMjP7dN3qNN/rrlGjA9CFN9xgz9FAahQk7IZv5M4T4fQIzNSq9ZthX7xRn0Wi+ipoiKJJ/LTOmY2JyCM6vlkvmzFaJao2RpGZsfEJ7Dv6jNkZChQszn86GqSGtBnBK/309fc2WidB7ngTU02QWW3WkTPFlvuaU+7WVCgb29PdQkKKMg9uVlIUNhdsBSWcCXX35ZKpV2d3dpt7q/v5/P5+/cubOxsdHb2/tf/pf/5fPnz69cuRIMBlX6ABVkbW0NJwkv+8GDB6+FWTufq80QKSlCwTOmqqU8yJlj5jgXo2mpzbltsSzMQMF8NdIG6e3tpdrZpEUrIczwFovFtbU1NbFk4skJqNfrLLpGo8FmCXRsATJODF3/8hXWr4gMPWZI6xxQM8tpTlNz5R8cHNCEhVJRqmCTySSp50AgcPPmzc7Ozrm5Obr5sngkkptOp8PhMET97e1tOiaIexAKhX70ox+tra11dXUNDQ1tbm5ub28fHBz4/X5kxtLpdLFYxPrTFj2Xy42MjNRqtYcPHyLrhfxNoVC4cePG2toaCXSXyzU5OclsjkQim5ubRA/xeLxer2PpwPVeKVhx0eEcz8uEkyaXWZZXCQenCIYMsRXomWrXRFVy1eVOmq69LuRMoLV8LoE/bPDYPjUzNadgG/FrJ4PK+RnTyza9b1Aat9tNm1BTxiSVSmGh2M4ZOvoc1Wq1oaEh+q8jFibBeCuRJfkUDZfOr5JLCyW7THKiZTjc/kDszWzczJCqAlacYlY4gRSI5fHxcf/LA5DkN7/5zV/+5V9Ct33+/Pk777xDFcLu7i4i7wJFTYDY4rq09EUuKjBxflG/UcWHVGhY+xeNw+m308liv+nkouhYvzfN/ZslkJw+0/T0tNfrhSWJKw3WUalUent7k8nkyMiIx+PZ29tjlrZMg1sbecv/6jffMLdMzqN1QB+ROKzTwUFYIJ1Ob29vUxG+t7e3sLAQiUSazSaV4o1G4/j4OBAIXL169csvv9zZ2Tk4OIhGo9Vq9fHjx9PT0yQ9I5EIgVsqlaKVTLlc3tzcpJx0a2urp6eHak6qEzc3N9fW1gCLl5aWgsEgnReazebExMTQ0BCZ2f39/aWlpd7e3i+++CKdTl+5cuXhw4cQUR8/ftzT00OjmcXFxZ6enkAgMDw87PV6QWYUy/+NHSRsTXUnNvlyuSz/1HydSqmZhCd9gJlt9lRlNcrg6qWztmV/Lz+hdXUJtOtP8qyd6ND3ckSjUWIIiNiCQeAYdHd3ezweLB2kTGoxgMJICNPQGU9KPRZ0fiXezQUistpFDPTvfpiK4aqoNjP/nZ2dkUhkamqKSBQDzdo8PDyMRqMw9sg30L8UZ6hYLLLJ+Xy+L7/88uc//znRQzgc7unpwV4jV2RGGN+vpyKnWPWxig/wQNtc8azVzGy/nYgDal7dSSg07aP6T1objOli9/X1ofwsp1azCLcaiST2VPWZswZHF3Wqtpm//MYlQuLaaazFqkGQkIARCotV/DMwMBAMBmltMD4+TjTKjCmXy11dXaOjowsLC1S4go7t7++Pj49fvXr1/v37i4uLdA/gzQ0NDdHAAnL06upqtVqNRqNcCG2weDw+Nja2vr5Obcve3t7Tp0/39/dHR0eDwSB6qn6/PxgMPnv2zOfzDQ8Pb2xsjI2NJRPJ+YX5Z8+esfesrKz09PSQAr19+3Y0Gi0UCqDkPp8vlUo9fvxYLc00aoy7s4zFRNZMI+ikfJl0KEGcMspSQgBdNeeTWclqWlvFv7wvTSz8XESuv4mkvsVzVCEGjEsi0ZQ1h09pza2LWkZpbjnXANuDyRYnaWZWPTiJWeZJLO+ehcTnwQrlUmmzIXLERtO5AnCpv78fPj4eKGXNRFFmOwsUhaDMy1LTqAUsTm1b8bt5OnXwcN7zazFDTJEcRda41dFolNDT4/EEAoFIJHLt2rWZmRm8Y9JiUFHxsovFYjgcfvLkydDQUGdnZ61Wq1ard+/eLRQK29vbmUzmn/7TfxoOh//pP/2nFOZ1dXWl02nSQpAQ6DWF+Tanq6l8zZ9AnzC7ZlLa5AJqxHhZVtmeNbtMg9X5rTak2RPDiv8oIDBhDXP8LRjTeXDmo6MjwnRhvyaUxyd5dsZWRfMqZQBQlTIfz2UCjNZmoGAXYJp29WpkzAGz+Zv47iJcSX63vul8VFjicn7p2NJsNtfX10kAbmxs4PxSXqhY+/T0FKIIpnN3d7dSqVALOzY2duvWrUePHlUqldnZWbQKaRh2eHgYi8Ww++l0muWKBtP4+PgXX3yxtbUVCoWePn2ay+VCoVAikbh9+3YwGFxYWOju7p6amkKucGFhIZPJnJycECTm83koq8hFDg0N8Sxq1aHndZbJ8q8SUBbGaqYi2bEsrMr0QBX1m1cUfm2G285m5NZhTgIhyyps0eRj3pt2k7ll9Qxqf6jAwYxAtUI0oYUA4tW+bkt4zmxCdhfdHnkXHNJsNksNFKEoehonJyeFQgG8y+/3q7cGmS7WEqxWC+vXIzgXCCFdx3c+NA0kSAA9oL+/Hxqr3+/v6up65513rl27NjY2Bp93YGCAvUSj1NPTk0qlurq63nrrLSAgv9//9ttv//znP8/n848fP/7iiy9oSP0v/+W//Of//J8zqqFQKJ/P09sb/f6ennOXjnvjwZlRooFqRrHfW+xji02kr/BJp7tjEjTNpdT57e5oFn+qQMScePxgqtZZOLXTw9XBxzDTCmQt8Foge19fnzN5yLaB3qlprM3rmtxozR9tFWpGqHOS5v1d/4KWNw0Hk89cRJTZ39+/du3aZ599FggE3n77bYDp5eVlhd7sMOwt5m1lMpmBgQFIF/v7+7Ozs+znvN3Nzc2lpaVYLLa/v7+9vV0qlfr7+9HwHRoaevz48c7ODskWj8dzfHwcCoVevHhBAQ5eUr1ePzk5CQaDv/71r3d3dw8ODj799FMcdsghkUgkkUjQH3NpaYk26pFIZHFxUeJWzH6gCRMzNWMRfja756BZYYVs5ow0G5qYZlr1e+bacKZZZHzNKWK9Gpa6csUmqA3QKc+aBWmGXerC11JASm6I5Rlhjp0QPx8bGBgQcUU0UN1YewkUs0bDHIHLROjMQ5yAk5OTtbU1og1kDHp7e3d3d2dnZ4+Pj/P5PDOK9m9ITgpWwtcmsjF592Ypc8f3BH0AQ4NjIIsaCATQvKc6I5lMTkxMQIG1ylPxCiFloTFJ1QJrBOw79fL4h//wHy4uLj5//vyXv/zlf//f//f/4l/8i8HBwdXV1Xq9Ttceho5NzrSq6vMpBWp+bw6R9XZkl51Nl8wA0dq/zZ7lnb+fRTTDODMWMZPbOo/T6JuH8zdWoalZZKhb3dvb29nZASU2z8OKo9RZsg1WL2OL0cTR1dUFrJdKpU5PT4ExONA+/B1vqaWPRixAOk5BsbU1BQIByLn9/f1LS0uS8wegwDKmUqnBwcFSqTQ3NxeJRAjMaY0IsODxeCiKOTg4qNfrOzs7Xq83GAwSjBOY1Ot1Gqh3dHSsrKzQbpIMMtBkNpsF0a/VaoFAgErFpaUlxi4QCKBSvbe3BzJeLpdx59nHvF5vLpdD2BqwOxAIwPYzjS/ohNOzNhNcpo2z9PvNKWUtdd6Zuk1/x97SmGAcTLOVlwlPmzLnst0YC6f1aWMZtWKVGbZkiUzyBrfBmL/uM2rErMZG1sfYCUAVkWuQQizB8uHh4crKCkq5cP6Oj4+p28SlgMJk6ovqthW/C9IFBeaL30sxp+mwY2HB3BEDSCQSkqKkaIA+fliHZDJJHU0sFoP+dHh4mEwmIRE3Gg2PxwM+CVvmZz/72ccff7yzswNx67e//W0oFIJKTDUya5kbs+aqVbXPYJrbsHWoCYbYqFbPDQurbXl0/X5ewdmywIlMWqjjKwk5cqIt8TzNc/iOxF4ma9YsRJRnjava398vs651AU9R5SaNRuPg4GBzc1P0Eg7YqOeYHew3sZSsjUh5KsoL0cwWoYJ3MzAw8NZbb1FGtb+//xd/8RfokRKXHR0dEW9+/fXXZKVjsdja2trW1haYBkjI+vp6PB7v6enhRsvlsjRh8/n8wMAAtNC+vj760dDlgFaKlUoFFVapea2vr7PLSQYB4D8Wi+FoxONxujJSZTQ0NBQIBAqFQrVaTSQSy8vLpVKpXq+zv7WUdTUHygrnyQSI1OlsK+WclMpRmOcxN0UMgdWAzaJRWiieYGjB2c6aLlNeh1ui34JFdSC2tbS5hSSy6rSAVbVMXSudRt1uN+3f9vf3G40GPTJQcIdeZoKY5rOYt2Elw1WWab0XxHLVoh6fGjoaNFhUogQQrays9Pf3gzxqPK026tQZMESm+IluzFR302ElhE0WjfWkSirQRArGdDAYNBPCu7u7e3t7IFQ9PT1bW1tssVBi+JcepwMDA1AUPB4PXhHoc29vb6PRGBoaorEqk59yB0Tba7Ua4SxLG6UX4NTe3l6MOGsf8pWcAOyD2+0OhUIma5goDQYBKv7YLxx2AW5aBWZId1HOsNuQmQPJBAcns2di3NZasxKPpqPNO4UfoYIDKZ1pTfX09ODqUeJniv/IMdJeLiarJdRnpqPwSrPZLJ1oqH7K5/MW9NRsNs+XlqmVZeo8cXY4QCpsAxYAc2SkJiYmQHDW19fVu0FtwDgJZYRYopWVFV4SkoO4A52dnaD1kHgGBwdNtgOAIJU8VBDRdoEf8CN2d3dFiK5Wq7FYrFqtdnd3v/3220+ePCmVSiTK6Qm9v7/f19e3tbVFLxt8CkgCase5s7Mj8oBlC6zwwkw+aHxZ/Dh0HW90CBxow29r4+0q39LmM5qLwmGtEFVpwzYOIy/aqZtzdnamNqaHh4dXr16dmpoCgcWDQISBdhY0gTZ1li9/OB0xVWOB8IZCobW1NZ7xfNK/bGqBwmKz2WypS26VqLX3+Jx4kQ6pNjq9qovmEj5TOBwm6ceiozMA94BzQ98MEA+FLCqTYZ+GiocziyYR9XjZbFYPyFIlgN7f3/+TP/mTTz/9FH4UKsd49/gftPSWGerr6wNKIkQmxdrV1RWJRLLZLFXK9JMFX0I4nnhFuYc3Y5t0f6tsg1Ezy2ouOkzmsdktmvFnU6FJIWtW9lCvDFFlkSOttBM7kFrwCBp1IjNOaqwaqmWzWUilzlKg8zBZngIPr95oWgb7+/u1Wg2fml0CSJdHhQAnkVwgNlM8iGD87OwMehDOtc/n6+npmZiYqNfryWSSkBOFPKQ56vV6qVSiOW9fX98777yztLSUzWaTySSaHkhXz83NdXd3p9PpcrlMVU4ikQBuazQayWQyl8utrKwAmGiAKKuhKTsFu1xrdna2o6Pj66+/7u3tHRsb29zcZN63tNcmldhSYhoeHqYjsOkWvcHR/oum9+Hk8whSlEiTJHLEFVXQKjFYJYv0gBa9wWmzmAac3Mzjd3R0FIvFYDAYj8enpqY++OCD/v5+mD/Icgp2wP+1Hs3ih7TPdjpVASC3ZTIZMh8zMzOUsNI2qFAoAF63BC5MS23dmHTd9Fqt/tEtXxNJJ3Mbsz6jBsS9vb39/f2jo6Msn8HBQTowSHkcYJAogToAynGxKd3d3YODg3hLHo+nr68POQumInthMBgEF6Klg6YofrTL5frhD3/4v/wv/8u7775Lu4POzk6Q8d7eXpQxRH2JRqPxeBwQBs+MlkxdXV0zMzOA6V6vF8E/uCJMkjYaT85fKnd9+vtfNCFgonDRXtmwrbfAD2K20Abe/AD+XLPZhGpMHb9J5vH7/UNDQ/ALTVgY/Jbo0Nx61RnSWRuh6xJoEoIUCoWOjo6FhQUz96hPnnvWvG+rxozTscMwb7hFYdyowyAEEwqFBgcHx8bGPB4Pu1OhUMAccwBE8KZRu97f38/lci6Xa3Nzk8bvsVhsY2Pj4OCgVqvRtYD54ff76/X6nTt3IHQ/efIEDT/0V4PB4M7ODk0uaGVUqVQ6OzvpJQYXW8hRT08P3Y/QAj46Onry5Ek6nfZ4POoDAMCSe3mQUr8IgGsDuXJybuD7UqRrqVzT5pBbjd3RuzDxBEGuzAELXjBhFqdQmXljxFsgLebH/H4/7u34+HgqlSqVSlSxoswpqgbyb9bqeq1DtcIcCKRpY6hWq4VCgamulY9D6hR+kqVuuT3wS3IYVpHORTfm/GVLtQ1Z6r6+PoSqfT4fAb55dc0BMrrgNmY1Jvi7jBdwJ5YO3flSqcTeTAUKcSGGhi3B6/VixchP4kUyYv39/fF4nA2D1eF2u30+Hx4SATeWq1qt+v1+JtXg4CBwOWG0KWxgDaCiOtO6mfp5Hb9/dHd3g7VSGEUvOj6MLoeT0UyDZr0Ik2WEFgcoAsx9q9wJzSxMHBuDSb0Xw1oQGXPSuXAseIfibeAXkCszb4lzeb6EOR1oKTW7NNZSix0adljI4OHhIbAvzd8ePnxIe1y682KqSGKgZkfBt8vl+qM/+qOHDx9ubW2RrV5dXeWebt++vbS0BNYZi8Xcbncmk2F7CAQCYnoMDQ0B6BBKp9Ppr7/+Gjg7EAgMDg7Ozc2Bv6NPUiqVPvrwI+Zlf3//48ePt7e3g8FgMplEqonX43l5QFGS97G5uclS39/ft+BLa9z1wlSUwWImanH63S2puDpeaY5N2Mqs7uvq6qrX616vFzU1Jp+wZkVCuiUl4tD0UmdlExKRLykcwFljTTLWzLbTHl7xcjQanZycZBYSewUCAchhnZ2d+Xy+WCwmEonT09NsNotXSPrBih/NFWvSHAW5UCoWDAabzSaCBMCmznFjXfn9foBOkGL6vZlG3+/3k46jGxaogtTxCeDa99W2XH5zAsgLE28HNpjP5wuFQjwaL4XFgmInn+GGm80mjaZwvbHU4mISzjKrWc64vdJc1s2YEgV7e3vb29toOlcqFR55cnJSpc+mZ6okNn2w+CU5pGQyifomf1X7VqV5xV9QOkGJR8suA+N0vGSbOLtkSJGDsEmL3TR2qFwosnQerGL0wXkiRslZJAVApFCJCY+CW2dnJ40DuStqzc10pYVsKCZWshDQSQtW16WpwO9Yt+KNqQ9AG5lExjcejyv1B+LD1g2XHtJ4sVik7RCCvKurqzMzM/l8fnR0FJ+XYK1ardLxi3ZcjUbj5s2b9+7dC4fD29vbnZ2dN2/erNVqLOaDgwNSRl999ZUqI4DGKM1CtI8M+OHROVr66aefRqNR1D/cbnehUOjr65uenn7w4IHX652ent7d3aW7Ixwp3ATyudaLt95xG8LZ93s4YVOl9YROwmNBygfEkzsfHBzkY5YEAVPzIsqKfOo2MkzOgwp+ZsjBwUE6nb59+3YymWS9Kcnu8/nUiwfd2gcPHrzumEhkCtcyHo+fnZ3RN8/ZZECcS9w9/mXmI67L2zSNdbVaJd0yNjbGQjo+Ps7lcl6vd2VlxcwoXrTFtuHzsUTVn9vtdhOtyo1V/SRABzZa5FdMD1Q8gTNm5xC54RgLk4NhUvFU8UHKEcgbt53+3LhuyKpwJ3oEnHdMiTkfSNMlEolsNsv9U6dG6xL8QvY8IgOMhjMHaFJjOy7ODegr3AkbsBkU4jewBKyEvOl6IyeLA4eTCzqhS9C7FYYbRVU6YSAQoCi6VqtpCxHejVK8RZ8VE0a2Wy/OapTzzV9NbjlzVxJOsAKcrD5prbKTLCws3Lhxg2zm5uYmfqgIoSyJaDS6uLhYr9d3d3eLxeL+/v7z58/p50JLrVwuR5766OjoxYsXHR0dFH8Hg8HV1VXCOrZfr9fr9/ufPXsGPDc6Ovr48WO2GUq8ANoajQYwC4ZbqRjYKVtbWzQl6OzsrFQq2Ww2k8lUq1VKJMRBvuRheZotxaxNWY/v6xA/BM+CQmHkithmyNrBt9Xui8UBj1NBsymB0kY76ZXjIPYefDKfz/eP/tE/GhkZsZKNp6eng4ODmEiMgtvtjsVi6hV9+evKqTw9PS0Wi+jGkV+xpjv5XuLl3t7eSqUCMuDxeBD2In430RicI6j95XJ5fX0dwA3ozCRcv8FBfohaNZ/PF4vFBgYGyEvzZsWuYTAxQ3AzeATCf2TfdWCzME+mQKBJtjF16k3P+hxT7vpme8jlchMTE41GIxaL4bN7PB42Ej0Cv8frN9FCpezk+cF/XVtbq1QqtMPGMqgETCbIxIgsZ+jsVV4R6Ci+I6xN3SdDRxJVZzNlGMrlshK5TFH1AjUZI0rOTU1NqVjm8PAwHA5XKhXSZmrnQg6AV6ZrWRGqmXmSL29tKozSN5ukujqpPkcaK+ahzClWgEKVcDi8vLysoi8gfLfbjTYVOna8nnA4nM1mgZb8fn8qlVpdXaU1F1KWHo/nxo0bi4uLhUKBGUDlq8fjobkXjXGR98aJQ081l8sh8pBIJIaHh1Fn9fl8s7OzcNe3trbeeustzFalUikUCqlUant7e2Bg4Pr16z6fb319/fj4uFqt1mq1nZ0dItC+vr7+/n7cjUsuv47v+7jonJrH0WgUehwRcW9vL8K7IF+w4swvqgyPl6642IQddAnNB7OuzJnkMb8FzIoB+lt/628lk8lwOCyZQyGSNAaiM1GhUMjlcsVi8c0GB+NCXssU0Lho3EAVaKwMVZZ13mg0KNEyPx8MBnt7e4nhyIuQg4pGozRU7fgOB4UILpcrFotRlyj9Qsm3YkH4V1K0okhhr/V7sy7D9Kwt62ypHupj5656f19fx7lixPr6eiwWC4fDBwcHiUSCy5HK0/3jXeIrWEUuqOvIWFcqlVwut7e3Rwwhjp1iYh4E5q7OI0PW9ar+3fxQrVYxI6Z2AlizmYDVV8yK02AwaPZEV7s4c+jItXo8ntXV1a2trYGBgUAggHuHki3bqrw3CvQIPkwtPJNfaFUP6NIWIeJ8tepzBKSQ4Uxvular/Y5R29nV0fm70tJYLDY3N7e8vEzfQn6g1QvcIHRCMKw0YOQleTyeg4OD5eXlbDZ77do1dqRwOAwUTotxv99/dHQEf45KLawGLQ5oOdzR0bG6uopgU0dHx+TkZHd396efftrb20uxwNraGtmSP/qjP1pdXYWZR5D+4sWLk5MTWkGXy+WBgYHl5WV2QpLC6gUMLMsPllyGhaU6weg2tXltysSdJYsKZoE7enp68HcikcjVq1cTiUQulxsdHSW/f3p6evfuXZo5LS4uakHix0mUHAlHzm/2CTTRTCsL5HwKqwIeSg8LeGZm5v3330+n00hViK/d29uLmW40GrQnXl1dJbcMFAvmSBtPk5WsMjnJL4gFjOa42ROH7AsgLHst+Wq+sr+///Tp03Q6PTw8nMvlKHaFm4xY2MnJid/vHxgYwN9vNBpUl9FeAx+TNqEgJ2pF1PGqA5uIAbpx48bw8PDi4iLtL8wkocQuRPbg96JjQawGQWafZpQg5+kqYm6Yds3yVam4qdVq5+qg++fxtNfrBTcnwaDuJxIgNJtP8hsSEhL4xTmF331wcIA6nTaS/f19GWW2K7xv5cYsnnXXy0c2W/AQDfAxYgtohVLS0Jam7leMrQS+lSR0ppFMVS+ztkCAuNvtnpubW1pa8vv90WgU4BfWudLy4XCYuIdNSBrrGM+WdsDJLjedpG88a7MtrHx7cAOzyvPcUr/8kdTZ/Pz8wMDA+Pj47u4uyDpM/v39faBh6Qj39fXlcjlWC80KqB8Lh8MUs/j9/kKh0Nvbu7GxQQ4aA0TY8uTJE24a15tMQrVaRb1P+oS7u7sbGxvQ/vr7++v1eqFQQPX0Rz/6Ee02IN5+9dVXeq719XXMxMTEBDU4THSv1wushsv2ykXYEuVwdnW5zKEIRich9AYOxnutVCoffvjhzZs3h4eHg8HgJ598srW1FYlEaEp05cqV1dVVNItBdebm5ijBp+xbzVsvQzixGN/OngmWZ82avHHjxujoKOtcXxd0zjQDAxFdF9+WTDXfcg4dG1JXVxfpQQJwvSa1IyAwUizMnu1yuUqlUnd3N7U5zJlwOByLxbLZ7Pr6OpM2EonQW87tdm9sbJTL5Wg0SrYtmUzSsjmfz+OXITxC5dclOfUaduwLuX6Tq26yO/SDANC+vj6kQiQ7pY/Jb5C88t7eniAvSV5oXqEXRow1ODiIpNfR8bm/zHBR8s4hJqiV4jNLVMgBkk5UT95KpbKzs5PNZre2tgi7g8GgAinrFWvJmKvg7GX2TxqTQmKJLdhaUFAZHBykmzB2XzvWN6vp7Dz617yyxEmEKYPiOlcupowvRiIRrBPzBC+bViqZTAYHn98DrRCF4De0rLpsmaY2IdbfNQqBQqBXbvnhVpn/wcHB+Ph4o9GAHYXvSSa9p6dnZGQEJ05ictSg47bkcjkeDwbewsICm1UulwuHw0dHR6FQaGhoiEfa29sLBoMUFlKzDtdiZ2eHCIt0PEWJhUIhk8lQsI6xZknvvzzeeuutjz766OHDhz6fb3x8fGNj4/DwUF3HeEC6RAvsw7MTjm9mJJwWrY06zBsczjo3YIR0On3t2rV4PH5ycpJOp0Oh0MjICLASslZgQZ2dnUNDQ2+//fazZ88ODg4ymUwoFLp37x7EamGUzobIooJcxDnTh51/5StUUY2NjV29elUhodIsGBfS7ngDQ0ND8DdyuRx9iiF4kZF3pnbJOuB1ylKLlIq3JTHrSqWSSCSguK2urpJoOT09RQUM571er4+Pj+/v7w8NDSETCg0cnjiLc29vz+v1RqPRWq1GCzpspeixppvW/gACYg1vb2+rul2mWX60KcFh2m6sEvwHJO4EnkilFt+QmizzjZtthlQXTriA7WMtw/rAf+f80msUdcE5K8SJBveAJoRrj2sFrZAieIsfZU4tKwnEz8cvZwvWmaUKazAcDmMTTdReBt2SsTV9ZJNxYWmPmPi7KRFlEt5TqRSV/V6vF/lP9t3Qy2N7e5vsq9vtxq+HH8zZTNzcpAw5cUhznH+PDULrGilqis7pLOqlwYTb7Salw8wbHR1lO71y5Uqz2aQpYl9fn9frxeaSMEU5mvoxSIsEJoFAgP5eONTPnz+Hv0HRAaScRqPx2WefUf1IEVowGNze3t7c3Hz69Kkoa9FotFQqMYiNRmNjY+MP//APQRhDoRBRcKVSIXDDUpNV4DMej6dWq1E1Q33tRV5zm/9quN8sqWiqDfDOBgYGkskkKFAmk/m7f/fvTk9P+/3+QCAAZYUU2e3bt1lpaF1evXr1L/7iL9577735+flCobCwsNBsNlm3phiIqthNRqB5J86jjQz66enp7OxsIBBgoUoITEketeBhEsM3gJddqVRAFSk1dDrXqBqASKiSy6To4XNFIhFoPGA+Ho+HHptCLUDtiJd3dnaQJTg6OnK5XPQP2tjYEPHJ6/VOTU2dnZ0NDw/39/fPzc1tb2+TklJ1vvCWVx40nEO7lZUsmNLylJVs1M9YVTmVWASzQ4qZM+SJnGpE6jREIQzNBygWl702SQtgg9y8xQaxHg3TsbOzg6sk6yEvQRwYk61xmWx2/8tHFj9PxC232y24w1mAZpppqf1JoM1UC7DUfjTrTCa+mRUkHnK5XMg+I17EqLpcritXruDCq9Mu85n7tHTxtGfAC7jQs5ZmK9/n56OjI16eqYGgLwMFBgIB/IvDw8Px8XFKCll4y8vLYB1EOmBV1FDRFQz7gqYSefbr16+vrq6SFfR6vYVCgQX57rvv1uv19fV1ffHs7Axx3oWFhdnZ2XK5vLi4SOmjlHcODw+x2qh/xGKx9fX1VCrFKp2bm/O+PCCiCMPa29ujGAFXnUyI9DScVdcXgRsmvGv22mjDxjXPqSnF5kTndSQXQqHQu+++e+vWLSbK0NAQm7kcCooCdPT19R0eHv7sZz8Du3///fcfPXpEyhEJR4wLZk5FjJbeXkvaojJ7si/cEvzLZDLJopLmH2IytETi7WBlID9Qf8QZstksKWWyFyxyFIIwamp8Z2q2SEBKBSB0gdnf3/f5fB6PB9ycjm643oVCgRgRkhKkYDJCkER7e3vpIRuLxcbGxgYHB8Gyl5aWuDdeEOgccxs2Akp+5ljhkBIb3bp1a29vj4wT7hGgqmBWOXfMQK1B0Hzh8vowt23y6E3UVYGg1R6eDV40eSFjnZ2dfr+/VCpJq9MEYeRfK+mHJeJPPAt3i2YyXuTS0hK+Jz41w8IKBaNTuRY0oYGBgVqthiEmBHe//JeaSan6aXhN6hsjoE6S5nQlEeX3+2FDIVNhmkXzMJkqxEy8NcYQOJcRY9NVtQHzkP9K/IucCr6LKc9pOWSwXXG8nPj172HWzHjGQqzelq19eM65uTne4t7e3ujo6NnZWSaTwcojSjI1NfXkyZPp6enNzc1EIjE+Pv7w4UMIHiSjcrlcOp2emZkplUpLS0sgcZFIhKbjIyMjS0tL6AkwcelzQVcBpmwgEJiZmYGGdXp6KkIV/+bz+Uajsb29HY1GcUBqtdrm5qbaCKFPouEIBoOEqALULD0wS1famYL7jpwQlT+RnwmHw+CqiKUMDQ3RYCEYDLL3mLJwzr2Ee4PP0Gg0lpaWbty48fnnn4+Oji4uLkJbVDpL790KCFpuLcIuzdiT5CcF+jdv3nRWkKPFwZ4qnkCz2VS9lsvlIiCgNhWOFJZX428CNSbUCGmH7YGUDnQ3Xk21Wh0eHo7FYpjIUCh048aNWq2GMYKQt7Ozs7y8TE+WdDoNtZRuMggVUc6nUKxareIQAKMjNo15UusTvVPsy3/8H//Hg4OD2WxWHC/4o4pyeDosJsgpxeV401hMFFNJqZnKFe2J/1bTMn2S2yO1K5hbxGe8Adlry4WXLh3mjDQPJf6YYPrGsvGgHajphPHlKvyezcZsZi1mUde3PB9xFnXzGH1TCpHzm7k+82cmuTb1NsgeJkWqZOy4JCpUYY/VlmE1wxHtkTjX6kwrbWunsebtqxGdgqFvKir0YLoz0R6t6EmvXBXMhUJhYGAgEonQ0xPPVE3Xj46OYF/g4BSLRSmmiuQHRry9vX10dDQyMrK7u1ur1e7fv+9yuQKBwP379/f29rCzx8fHa2truL33799H1sPv94fD4UQicXx8/Jd/+Zcul0t959SYmVwlQtWbm5sLCwtut5t+Y06QyO/3Dw8PP3nyRDqoli/slLhz/um1BPWtQ5eTYsONGzdmZmbGxsZOT09jsdjk5OTU1JTiX7MFhnWojI2fU6nUzs4O/SoDgcD4+PjW1hZeBrDva3Xekk8NrMGzQ06gLxqG2/wKCjN4HEoDsGVC4JMucz6fJ6GEn04BhVkZYZoeOZISVpR4IdK7eEY0mfv8888DgQCWXW1NIKezFMlTAceh94sb1dfXNzw8PDg4iK9H8Tp0kXw+D4BAlyLCf3YaPfvx8TFDXSgUSqXSN4qXL+0jgwbdW2+BUW00GtykjPXk5CTlxJTM4VhY4VT7g9liBcqsaIyIWqRSnSBautxqi3wmHSWSuhDVNzc3m80mg3x8fNxoNILBIHQa3QleKmaUbJYKfICnsXSioPS9tPUkQhW8mtw+K8uqiWrKPJk6vdZUN3l75tJWtpZYn2p1yc+J6KxkgGmsPR6PngIKU/tVhpnm2U1j/buiGO5PillSLHTmo7Vg+vv78Xr29/czmQwLD50mRp8qILMelK64Q0NDhJlEWPv7++l0mi4E9G1BWQmY++Tk5KOPPqIRQTAYvHLlCoUJ6XQatgA562Aw+Gd/9mdXrlx59uwZzBCVrqbT6bOzs9nZ2c3NzUqlgnWjBTvbPlGYnosljb34GytNNA9J3xKPx+Nxn8/HjphOp+PxeDQa5Y0ALFyGU8zC8Hq9CHxDwLh69SrJhmq1Cjr/uhomZuKL9e9yueBTsmUq1cPnqYYQtiCGGXskRvb09JSGxUATKhAwu/Ao5rOEe0j3idtAzTp+6OHhYS6XY+pzWgkCY1kAJbBZ4XDYRGaF14GSs2UCrxH00AUJpNvlck1NTZ2enqLQqyMWiyFexnyDmoL1oQq82Wxubm6aYwvMgq3UnSwvL7O47t69Cz6WTqdv3rx55cqVlm/f/O9FlZYKUCgKZxwoEqGiT7itVQPNgd0Ap0I84NGjR8vLyxABEGgNBoM47F6v1+S5AsEDgGhDgqDi8XhQ5xAMUqlUfD4fmJilrd/f19/T+zt+hZww7eUSJmSWyskw2TtWh3v5aiAHTDCV6SOrInaNpdGoQxld5EVbVoabmx+BvslUMeP4c4Kq/BGuCmJCvl5v0dx2CN/IrdPlNpFIuFyufD6v9khyW4CJeaOBQGBubk6+HvSynZ0dOB77+/sjIyM0Isrn8z6fb2ZmZmJiIpvNMij9/f3Pnj3b3d0tl8vxeJweoOVyOZ1Og/VMTExUKhU2Em1K+KE7Ozu09GVmYKYTiQQCCOYKOTw8RE9VKrTokwhaMbtXtEweSrzRGZY6UweaXvyAljFbN68tHo+Pj48nk0lSUiohk7Nw0S6ti5KSOj09feedd8rlMozGhw8f5vP5/v7+oaEhggk8QZOH6+wzqYHCgMpYoxuO08GNWbn+Wq326NEjim7Zh6QeGQ6HkXZijTWbTSWTiX4E+OpsSgqp6yDTWhRsbVR6g9CuAQRJfqhKOBqNUl2FZAL0NUls6yn4ZaVSQUwGkkmtVmNnpTZyb29vbm6uWCyCOR4dHSFxR+0YQC0VA81mc3t7m64aOJWJREIrEw4rxc1KYJIoIvUHVfHs7OzZs2cPHjyAzDc7O3vlypWJiQmCdGnIgQ3KhJmJZVklXDF2MuUwIdsx6+S34UQrzUONAiyAw8PDTCbz+PFjrBuIMCwIKmxNa8hfuTQcRGwo1YAwlyWYfHh4ODQ0xHXZIM1+Aqdn5ytaT2Q6EFqqpBO0hNUSHtLB7u4uURT3Jl0zPiZmGrZRqnByWWTuhTvzr2S2GFtiKXwIWVQFx2g7s1kiRWmml8+vKLqrXgDvwOx1rWy1aWgwbYDXgUDg+fPnW1tbqFK43e5kMrm9vQ3Xvbe3d2ZmZm1traOjI51O47zncrn19fWzs7Nr1649e/asp6fn9u3bJycnKysrb731ViaTIaHc29uLfMfR0dG/+3f/Dmpqd3c3tBNQXRxhiqPOzs7o05hMJg8ODlZXV58+fVoul1n/mBXir+HhYZ/P12g0KNWh1JiaHRIvyBXqNUj+0Wx22fIwlY/0G82kNt46g0xtgnTD2Zbgt5GXewN/nzzVxMRE7eXx5ZdfEnA1Go3l5WX0tqxWZBcdQj8EEEu8Qpb08PBQcLC4B36/f2VlRVNZIge4CNBU1X+LuFiSDmRErShScSJetnwLzISiXZwGmHYknWKxGBQ9MHSSGZSfgC0CR1pFLlDy/X4/aS6A3UAggBYNfkA+n+/r60OMDAY3LTiSyWQikWBxeb1eFfQymVWIQesW3KvZ2dlisQh/MZfLwSXgTZFR7O3tnZqaQlGzVCrhdjx+/JgXPT09PTExAeeKGWXqhFj0BqtvFn8CEweRV3WlCREQajx+/Hh8fNzn80GCRAfCTO6ZSTz1TdeQyo3VzMHDgyYk+cCOly58d9c3LsLB/stm4d+6wrorNWLm3UlYTfI+8js1N5hXJlVcRk95fowGSBGYstklx1lnJEUX0BL+xJ2DW1gIuyAU/qp2Ddhx/XC+6nBaTehTZVS4ACa5RMUplBsAd6yvr29tbZF7UfffgYGBdDo9NzcXj8dJ2rC75nI5QDEKz5aWluC007kcOWmcAgpAP/zww4WFhVqtFgqFUHSDxkC1C64Qte/s5xCqaKIRCARoVpBOp1nzbEIAgk+ePEmlUnRbhzhB7kjVDRpogmurwuUisNhK5lzkXDvNosTE0QSPx+Ojo6O0cQqFQuR5TH+/49IHhhX4iC0QkZ0vvvhieXk5EAiADlu1gtYTyV2VPIV0OVhRGqL9/f1sNjs+Po6NIGgjjYl6r6T0mX66CrApe6r0jHDVyUxa70KbgZk6w+tUqkrt1ZvN5s7OTiQSWVhYoLktHha7PuWUNJ0wCcs67cHBQSqVYhmvr69jj/b29qA6iCVFCQIGDqMzOTmJonR3d/ff/bt/d25uDnBDCUM5UJBDdnd3qaS9evUq6MFPf/rTO3fu9PT0jI+P49Gn02lNgJ/+9Ke43hsbGw8fPlxeXr5///6DBw8ikcjExMTY2Bi4k+aYpfYje23hYDjRlDhA3zIlwCg23tnZuXfvXn9/P0xZGDhg9zoPWVknz1WrA81+vUdcRhBw/BVeVu0lw13+1vlJTs9Ozr7pPi6fWkOqynXeFG/HFJwSg4ioTjPc4j5LpVJpBtwIcyO32ClqOmOqcVGDAugKBqhGMyrLxDsR21VJXez1+eaJ10NZOQKMAI48/MbGBsGmaZtAigmO+vr6FhcXGUqiOSA8WL1utzuXy42MjFy9erVYLL548aKrq+v69evVanV1dZUM++npaSaTyefzavaIdDVAHk9CMEWt48HBAZ3O33nnHfShlKPwer3vvPPOr3/9697e3kgkMjMz8+DBA8ApaCqk1yifGRgYoA2jObHYMMjR673SfYrCfxMRa2N55QJf5FOb1lZng/Lo9XojkUg8Hr958+bo6Ci0f4mdmi/ikiabF4oqwOHhIUQ0j8fz6aefirZlLl2rLkZPKksteoAyjSbngaoTMEolqymbQtl9YGBADhfONTm9fD5fr9fxicgmkdDr7u6mUwFRoDITOoM5GnTyFsyKuxqJRPx+v1hACP00m02zdQD7B36NufzkS66trU1OTlJFDXrG8gOD3traoqmK+tIlEokbN27ggQ4ODn788cdUA0Avwe2S48aEUTPsRCJBSTDakL/5zW8++eQTdICLxSKyAcT+tLvr6emZnp6+efPm4eHhxsbG2traxsbG3bt3FxYWuru7P/jgg9HRUeyCNRVNmpP5S+zXV199RUFDNBplveNg0rz0yy+/PDo6+tWvfnXlyhXqJChSF6XdEtawkniaydBnpT0raQdQCEyk76WhABf6BrLo6uzu+CZPaAo8SJMO10dEb0Bws00ME1gd9ZR1VIG+EmCSWKKaQSp6Zs2XvHsprJrLymz9jo3l0moWAXCPH6mIU2Jt54HFz3/+cyyUx+MJBoNo85P1JgKCloSfIgM0MTEBu4POTJjRYDAYfnnAD0eHj5lKqdjz588Zo4GBgZWVFfFP+/r6UB8mKqSH0MnJCbE/7ciuX7++s7NDryN6iZVKpdXV1Xw+X61WI5EIuSZiKJxHkkKIFKOmzVqKx+NMenQPeEnVatXn89HkqVarJZNJ4UdcUUH3wMCAjDVz0dQ2Myk0zC1JlWK2sHFocQUCgUQikUqlcKLHx8d5ZEKNUCg0OjpKgaLZ0cPKVl90mPgg65O5EolEotEo/f2uXLkCYx10QvJ1zn1FPFAJVhDSwrqRdI56TRB7xmIxsKNms7m1tQXwolxib28v9KFarYb+BssDq222ZgdwhLPIYpaYGe4Fk5ZOb4ODgwBHIF3hcBiiG7ry5XKZtshgX2LCqDhNy9gUwBToiZoj0m5sJKQ0sOAUa4C5pdNpxH550bdu3UJtFa9KaXxqEQHETSmfs7OzsbGxycnJX/3qV7/97W/feuutq1evptNpahFgsMAbw3JhxTDiLpdrZGTkvffe+0/+k/8klUoVi0W3233//n0EHmhihwNoNdbg3ng17K937twpFoulUgkqPXkCyHm0EEkmk6lUivq9c455T29f/3kIpY1T7rAwa/4KsZJtnoUGTnUe8necv/pvtAM7Ok+OT9z959U65IFZ7BDbldjUWxOui1tNaAKxGv6rCYNoXZiBhUlq5P5x/BG94bvqrcr8N7tVUBmAZpzkZ01pbPnBPKNaSDP44i+aO+g3LhGWXoZYiQjumwjRGSIhdNfd3U0HkPn5eYoOSPozIlIwYQU2m01WFLoKw8PDHR0dmUzm6OgIBj61juyHqKLk8/lQKETTL6IhZKzffffdvb29tbU1NvClpaXV1VXCHGpwSfVMTU2Fw+EnT55Q1FcoFEKhEHANlThut3t0dPTFixfqMcpQrKys4Ibz4GaWiXoKuQlMd9EElQowJdDIn5CDNdvOwy2VLI6+y1Mgl0EneBMge7Nadn0LPow0gOiz8/jxY6/Xu7S0xPZsNsl1nsepV2ml0RW60umYrnRra2s7OzviVmazWRAJghvMKCm+ZrNJiq9QKEgHB1qbRPuwzkQ8ujdcB+6ZPHM+n08kEuwTtIUNBoMYNQAxgY/MWPYbNiHVaKg7nak3X61W0+k0QZ4gTjx3BFfpVYRIzsnJyVtvvbW0tBSJRDCOkrd0u93QwyHhqn9pLBbr7e0tFov/7t/9u0Kh4PV66WCJZM3Q0FC1WiXgYH81U8E4DWAItVotkUhMTk5euXLlBz/4wYsXL/78z/98bGwsEolA9FaooYP8EDUpKnPb399/8eJFPp9nK+ISKKWoSOcbZ7z7m52Gx9FiMYFg68B3MVugAEZrwDX3jr9l6JfLZSQWEJwQ9xm6kbYuzDfrTiLA6ngFEAwnSt66prSMIcl5UdGZY2abGKdaDplMNifMOtEqgAz4mz4vqr4CF2HoBwcHkC+AEL9h2mHIJZelE7F0pXAthy6bzZbL5dnZ2VKp9Pnnn0vpSiVn7Ei4G5BMmRPcOgHOxsZGNBodGRn5/PPPtSx5Eig7tVqtWq3+8Ic//Pzzz1+8eKGGudvb20+fPuVJxsbGEG9ClLW7uzuTyTBGrNJAINBoNPL5PIOihjK8oc8++wxTi8i62sSI7GlCIoLtZBSENCnnJo4EsD6dzExRSrVGNNvUSswMSxQKhWKxGOKiSsS3R8lfeZh8cDYGn89Xq9Xeeeed1dXVubk57iqRSFBJZH3XrGNWnwpiNxB2JAcI3yhLAcUaGBjAWIMh5nI5qpzQtiWpqAQvghKIf6XTaY2n+dRqYmA2LYX/B7iRSqWApwmzoPrKF1PRnZgYXV1dNDGRVKziSJalz+fjBti24UGqVRWTzeVy+f3+SCQCPkAqJRqN0syTRqM8JiWUTqIkwvyqQtzZ2Xn69CkSuNPT0z/5yU/Ozs7W19eBR3h2cRLM96t+qmofOj4+vra2dv369eHh4UQi8eLFi1/96lc/+MEPdnd38Rb5onJrmBuU2kiWQD3EU2H8mcbf7KDHJyen5+E8cMHR4fmGxOYH9Gwx2yzggtWkJFvXt3lDkb75WP/LxiYgDBhZ2huJemtpi+Pl4FnTJlCZPS1ngjxcUtVqarMxk4EExKpqsYh6VmmFImy5L0Qk/BVMTGOOJSFDIzqgujowzt8o0XNzBGJcw9STFUgqDIuz+3y++fl5aHMiKpHDoXAWmFhQTiwW297eZrL29fVNTk4SRKA0zfbYaDRolYQcGhH04eHhs2fPUEobHBx8//33y+Xyw4cPx8fH0fHyeDw0UYYoA+xAr8VCoVCv16FPQUoBl19YWOjq6orH43AkBgcHM5mMy+Wq1+uoAPp8vtXVVVXrKmJVNkAZDFhoxIDE5tgssknWu+c9CSmzisE01TweD4xdYFZ1jviOxtqK71hLKIf9+Mc/xulLpVKffvppS+k4s3OjichjX4hL8EyhbzIHYEwTVqOyjSkfHx+neFfJBiareNPs1qInwVeT7QZaMasE+QqZtLm5ObyYer1OViccDrMmsW5cFw+RtQGaocyBPFZcOUlc4rsMDg6GQiGyIEpPQTXD6JA8BCyampoi+gYPxJjqnsnZANkxFTGOu7u7n376KYULoVDoRz/6EQEHLHvCR9lKS84CQ8DGw7V6e3sfP36skPftt9++ffv2/Pw8cYCcM3Yd6v4TicTm5ubW1tbExITX62VpdHR0kEjE+ggCdvWd2xTJs3R2dfa7fseYYv+TfqeWktmSmKdgEckjtj5/9nIHIvxNJBKlUglbREqTMIvSKh5ZDaQg+RAeUZ/B6mM/Q3CfyIZtgAdRRlFAB+1vZHytaEamkvIR5PeIsXhAPHSJQJjaI3DSdnd3GQRh98QKOHznHsY/+2f/TOGJySpnEp+LNhwdnwP5BnvP6/X+R//Rf9RoNObn51W8ANs0EAgsLCzQ24aCq4cPH0rRlF44qVTq6tWrn3/+eblcZlCY2fCg8cgondjc3GTBQIdYWlrq7OwslUrBYPDatWtIgBMtsjKhl4ZCIXrm4oCAoUPaxa1eXV1FDRLFUZxiIuuDg4O5ubmtrS3Whiq4lPJi6VJVTGwuEEDpRGUgzR71DCzzg3EwcWG1BcFsRSKR0dHRmZmZYDBITlm8i+9F0k+XJuDNZrPPnz//9a9/vbm5ef/+fd4UQQw0HtkC5ihBGXDB5ORkPB6XOL2ZLhc/hGspkft7kpXfNmXW+CCkpV5KEPIw2ezH2F9MPEsL8XQznIc1hGy6GaKiMPNN342XQ4qHwe6VTCYFGSt+132y2FA7MGmd7CtUMxaLRR4hFoudnp5+8MEHgUCA9igwYSTVr1JpTAOlH1TDn56e/uIXv6DKMZ1O/+2//beHhoYajQYCINyqignNfB2zjveirhGgooVCYXt7Ox6Ps11hgFArw0zoK8RbyWTyX/2rf7W0tHTr1i3VvJgpDQWaZCzq9TpxFXIfBO8YJpjRhFBMCXwamGPOYgVxq4m/fT4fNQdHL71sUtZ4h1DITSdXFU8MEZoNjLZofGIxkpcyiXQE05jm/f19cg+mJ07mg5EnM4+9Al5Xg2y1ZAI8EaKt1SfvTQtK+jnIzME1JDTBOzy/OsCKVTwtv+k8S9tzPvrqIc8Hnj17Nj09ffv27a+++kpeOdmbaDTqcrlu3LixsbHx6NEjepPX63W0LKBqra2t0SSQNYMsaqVSkQoBWSYGKJVKEbBwq9PT0wcHB19//TXljvF4HPH4qampUqm0vb0NBLG3tzc8POzxePDoof0xpj6fDzkhTAnLvlgsbm1t8XT0BpNkIGqrIBK4/xT4mI2lzd6Gpo8mbSMO8nLgU5YBNVlHyng4hUe+34Ni61gsdnJysrGxsb6+HgqF0BNX2soM8czegHt7e9RVRiIRMhba8tmTTM+IF62nMJ0DqF1maS+yHmQpCa5B8dgA+BP8MGa5U/IGUIX6QDM6Hh4exnBArWNxYolUNKEFZiEVZ2dnTFFsRMs+kB6Px+fz0X302rVr29vbSLLBdFaNtdIGOHSxWGx0dHRwcDASifT09HzxxRf0LB1+eVD8hWAp9Dg2MLKaYvKoGIS1LUkj9iTSjFgusm2UyweDwVwuhwNIFpHajXw+/+zZMxqGmIJZOpRwpl8Xnj73qWnMi8angR0rAX4zfLcOlj8EMDVx7ngJa4Cpsg2jdWG1DzUDVjYe6Z2qlEwQNruFDkIK1JVNvVmVm/DUFEAJfZZCk2l8zTljZiB1IGmAGAamlXs7OjoSQ5cHNxGzbzLgznmJrQmFQrxabktrb2hoaGVlpdFoVKtVrD7Fvm63O51Owy4oFov9/f0zMzOrq6ugS/V6PRgM+ny+xcVF+CcEVrOzs+T9IGzW6/WxsbGjoyPCQJROhRxRZSu53s3NTSk0NZvNSCRSLpdNdj1VMzhl9XodW4mgF8b64OCAojIkjNniCLql1071oNIjLHKWnFasRRFRG2bq+jiUNGdD1u+JgrsdxyW1+i5ztKzHwTANDg7Cg7x79+7AwEA4HE6lUvfu3Ts6OorH49vb26YgjiIAJFspmyasU5dxscqU/Zc1MSUMnf6BEDlpTRDYkknmA+iXNhoNNmD+qqwRz8WrJy1hjhuiP3hPHo+Hnp/qXUIEioGTsTa/DuhHwZs5JuxPZ2dnMH/effddrAxGELklIGa32721tXV0dETkND09HYlEkOptNBrFYnFhYeFf/st/SQHUe++9h1+JOgL7Fq8Mz1SbIiQZsiZm51yTOI/nTqJVfUVoJ9RsNjc2NoaGhrLZLPboT//0T0m/O0U7zaPRaNDRCT8Uf1CqcLzoRqMRiUTEr1cSSDPQmpn8lddnavn2GZpltVrN5/Mh0mJKDpiMUgB0NX0XgU+BLNG/mSfHbzNbGypHpc0Jaw5cLOFAs6WLZZc5v4r4rUOafPwMvidCCz6K3L7fzUiT9SFDQ8EINCy6jHONSqVy8+ZNeNMrKyvUYoXDYcQTms3mysqK1+sdGBh4+vSp6JbDw8MIvWezWVIxfr//0aNHjUZjenra5/OVy+VKpUKNBg4sk7XZbCJqur+/v7q6ygKj1Rb4Rnd3N2QP9gDAkMPDQ8Ru2EthR1F5bOqwcBIIhaenp9TF4MWrrZdaRuHWeb1eYh+ryazpWVt60GIyEeYzLfTOFNFbSvMm+GBZ3teSLjFhTbPQQ8XNpGpnZ2fX1tZU/idFJNXyyeOg1gO+MKxNlePzRTEo8PVUm6BF9f9Wdj89cZZrGMATCoNlhhYGRkqngH+a1LpRY1y4dunGtR/Fj+HHcGtOY2LiwngSN9oYXNDahCYtCEyZYabgOUOhcwI/vPKcd+jxOIuGtsPM+7zv89x/rvu6rzspcBhOCcDFGtJq6XBJNZHGtlotb1aoTIEk95xTjAysl42xs7PTbDb7/T6f7VR7v+ceGKS0ergB7XabhcpnKmuXPDCCq/bM6enp9vY2a+s+371794MPPlheXg6yic60ubn5+PHjhw8f+hwTRO/evTsYDFRZiC9CCXq9HpSfCQtZJaJL6eLzEKV3SiBCaUGYYOL4+Hh1dZXKwpWJM+xrY2PD4CTTJitljNgBcLAuHuCJeU/I5g5Ot9u1qw2iy+RGpaCKqjUoQ1c3yl32xh/n1ANUAqeP3awETOWYMXtbj77rqdVqPLS7V54jPXR7e3vuUqS3s6VlkwjywEkm20hJ3/4qrZ7y0HFyohMNrhn/FBoi4nmE5i9u+xdffFG2EpVgSqWZh9yHws7BwcGdO3cWFxfLsWNlzh6n5F9mZ2fn5+eF3tvb2zh8RCazGORu5rW0L2VPXWWkQgxZCUfkA3NrSj9fWqtKg8m4CHWJzb1KNTRosvrA+BteNfqhtL9l76lM7f3330eCdgeAmxXM+i+b1yutvVl4ua7cn06n88MP/7x37x9bW1tqs3ru5Shci5zmjTfeaDQab775pqZqxH4fKCgov6Uy+qscquAl/CkVgQOseYOoLVMx+UthpuBITFeRCcbQUGdDwDCTQRfYzs6OaSYAHPCCdAdVVIwi+m42m4PBQAKXrgrTT8C1kn1CvlmX4qR8rqQzLy0toa6SXnj27Nnjx493dnacT34O2+TGjRvvvffeRx99NDMz8+TJk6BParYsl0dD6sRuQekD7Dj8BAZ2dnY+//zzdPSwX4qrtAGsrtvt3r9//9dff719+/aNGze2trb0uHlMYTFljWeObXg8Uz8jTXOrfFLZ46MGdnJyQgcG4RUDj0AFSFoWJexVIXT9HuKLP4ufLPXx8bHMCSvB5gQxpf3Vp/koWU5mwkWfTuVZeI6dbRtLv8BWmA6k6EBJzKsbUtHFvvRIKqIoVgkW3SLwtAQxGj58iXIod+hKLhHDLX8uhzgIco+OjkDPfhiNRtJ8x89FV0QUI+gB6aeE506REPSNfKlYPma0QnKq9ONX1IJi8sZ5UaUaasVYu+/l1aoMlPbuVS3j5RVG96C8+PF4tgKVjBtWSJbgIrq9l+6AvxtZ57cqF8NYnzfL3Tblh9hTq9XqdDpWYSSm9PnZs2dra2v8N/uCtJvvyvL9kKi5kiJEWBUBDjvemB4JZgij8CsPazQaScump6fb7XaWU/JksUXhXUgCqnOGhRrzNjMz02q18DFqtRqihYKYcE9SLw1yeKI0guiC59tsNuv1etxGroERh1qUJYqjo6O33nqL2sH8/HyYhfpBmCTB3XA4XF9ff/Lkyccff6wRYXp6WvXScFEGNBpDfk5m5tgeHx//9NNP3333HeMlh2N88WHUGOkRDgYDHVhMCZPKpgs2IwNtLdeuXevsdRqzjVTtFhYWvAGHJ3bHht/d3UV1Z7wU6j07yjykX8l4mX4QJsnUufXkMi1TARBKyYDw9xKCubm5UOhihWXY5VkIlgARFY6AsMpBaP5X+UE8kbBXll8et3EDKLNUTE6RXDtCdpSbhlYk1w/x7KJi6VYmRK10N5SHHJM0dDQz9NzEyvCqSzFWSjRpR64MN4vGSooD4cRcOjV1/FUJ7SuOp+z6K411qEWVT6iIs0AVy9bEyhsq7NFyCVzauKuoUE3zOZhw8vpw24OF/e9b8bdw7biNaDLw3O12e3t7Ozm1czg1NbWwsKCc4vTqYeFUcvcqjQ+RZRjX+C4fVgbIXblyheK2Mlca5X1LrVZbWloSggm4ZNOqiGV0AytAsNOKregSBbhQpy3QrAMkYrfdqitJgOWnsIxokcbFdM15ZcJs8hKvg4MDkSAkRwdjxDZDP9UYxcHcu3dvb2+v1Wp9+OGH6gS81FnM9eLk9OXFJmGt4rREZBsbG99+++2PP/74+uuv//LLL8Tf9YBAcgVMQoTT01MNRFoNEKI4yHSXlLT3bre7fPOsfkBGbXZ2dnNzs9VqNRoNRIvr16/bw+4VfEmUirDYarUi9uvUUEPlCMXgMOirV69SpAC5eFLHx8cKxcEqMXyibOeuVtQUKtkPP6QllYvNjFDNGdFe5ww0fHF1tGX+cvxmjgYUUZBucwqAMhUBiTmjFkvTejFNJ/JAAfIrziH0+xSgg+IngSptXEXXn5gANYZr165VkFMvwY4IpbQmlY6S3Pf/X3+54n5KY13qWlRgkHIJZQFt3CCWhjil+QqSM04aLfuMyktqNpudTmd5ebl8LhU2SAmOX8rnK79oHC2pPCbPlAnGTVaPUsGn1WXag2xdV9Tq6qpDous3Rbmy+yZRSQn+VlyOb+HOUaz0Q5dlmYmJCQIpMsfRaESHAGRkw5QQpGqHWZ0ssrDFn2Jn7AX9b6Dt9D5E21oE3Ww20YeWl5d/++23drsd3c5oLbjscvm2VuLx8tn1+/3l5WW8FLEbEpiCIbGRzLuBOEsvvv/++8XFRdHr4uIi+yLDyDBG5ka4t76+/s0333Q6HQD0l19+OTk5eevWLQoH+IXRYOH/7t+/HxE3AWO5qysxHOfqu5zNubk5Y/nm5uasKIw0Bw2xz/w8JSL+AKxBhIc2rM2TqVK9c5ieZJs8yX5Q0JLKzM/Pa5MpzzXnFBgkjO9E1oBpRRcmWJAhyHV+HYoA/TMzM1JPG6zM6salGoKJR5js8PDQxfvHcgC0xxf1krLx5QJHhx+RkRW5kJsBmuhBCvmm3+9n3gdvJmSenp5ODaTT6chiUoVYXV1VuRamxWgGcrIeIJH4wjGbnJzsdDrkHSKqGXOQcdGj0SjsYI9QwRAxwK/IZy0+Z7K8udGR8ev9fr/VarkSJDYwmcPJVKVTq+zyCE8oBbHMuBMA7u3twRnTLOOpm3ElgykttW2UO6bgE+M4js9carhdhoVnQm6iOTnv0tISKrGgxtYXGclDSVdjnXe7XdhXKEoed2RUbY8w8NRhQJOxhg75ycnJyspK6ZPEL1RGS4iJVFDmmGSueeWE+HenTqlZHRK7GQ8PT3FycpKWdBTfcVGgdogi3nZwcIDWKb6+fv16lNPtNBS9oA27u7v+iv0ZHicgcWFhwYNI2YOv8mcKuZ64nhoyD5ozHz16lIo0NVEhZLvdpsP34MGDr776KvmH9tFHjx69ePHinXfeuXPnzsrKSjlf8e233xZO4UHTBvDCQPe24XDoXAwGA5V2KBkkXTVyZWUlHOfhcAiYVpfzuDVY23tR6Ys6sb9GjcTb5s4B0hyElFLIcEbwyAlNr7YdDkmIiktE0mlBhxEHAQfYInebGjo7O6tVTT+gD5eKIfg6KUBzIbkKBDSvrKPa6ghUmXMtO8ydYejUV0M0PDPCn3zyiQ+6aD8/fx4URrKBGGu7X3oudXKVmSEdpgTrGaJlujnkC7F0RuiC2/gcN9GR9lwz4Q2/0gZScOPrBAVRgE0a6AFrmcmABsnB/v4+AYSpqSlCr2UZnRKQCoCvqMyT5bcuJoT+8S8RnAMGN3Qbcx8yr5oYcegr4eVw1FbKqFEgqlB9o8qYoXBhlZZGqhI753X6368yFYi2idP19OlTOkfIvBpJHCEpp3UNh0NNFtj7wa8U6N0KBTEGSOaEk9D481Wv1xcWFigNQEtDDSxNcERzqAiYmdtoNLSxCQl3d3eVVUJnTq0y5BYGN6M28MkYVs8U/KLZMuPVI7AZuVcicDIAjN3XXntNf2ziUO3jPBytQcW9Xq83MTEhIA1jQW9xStloy0H8ZTbq85DxQMkcOa2V3fPXzz//vLGxcXBw8PXXX/d6PcgABkW9Xl9bW3v33XdXV1dFdo5w1KxOT0+XlpYyay3M+kwQ91cCeMKOVFy1jImNQmaIM2BDS6wVQGwwhXg2o4Ej/ez2IrTVzmO+lENEQuCXqD/m84OH2G88kGG7OCQpFzuMqV1nZpgyYKNxBseziiDmYEEx8SkVpJTKZFUCqdQzJSKCvMi7cyeWGdCYA2a1z67h008/tXJ9huzj1atX6/U6yT3uAmcFXhlif77bfWGhpHIqilHF5Q90YZE6oz+QrnRLdZ7FHXEVR4dnpaRoNcT6h8vMttp2mbw3MzPDGci27BjLhGnGZZXJXWwWvpE1StIzjTdjAc5swfTZZpKFufVCM8UNd6M0mgnzk68EUo9YDDtIWiTgUjn8wQEuC7AVcCaHfNxSy6q8ISEhY81yjUaj9fX158+fb25u/v7772tra7u7uysrK0wMelOmhYoICBilzRIIGzVei3UessfK6RjWGBmzrILRNMwwsSGLubi4uL+/jyud3lwBeKaeEy1wz+XFYpxutyuOLkMhQTQLqM7pDTR1U2o7PDys1+uiYz3fumQnJydDMwD4GgUA0TZ7yBtIDDLlIriw3TMQ1lEvCaCHh4e0fCMZkYcb1dA0Q0xMTDx9+pRIA+Ez9zMTGETih4eHIC9NyAwEFR0GAjyiE6ciQ+g+e5pRQEzDVGiOrDyAKCW7QLG6kHxFZmWJ+fx6PFZ0OV6e29YoEWWKriwt5QQ9EBop9NHoBuAVonqYuRmZBWMhUVQXGpca7jmwEZ4Wrdpg0flJPl1m7UkiuS6xF6McNbFMq0kEHJWxs2//7LPPUlfVlQdSYaTEDpABesSBDqA5jC++rUZ7WFK3243eq3QDcM7GMdNMM18n8RSYc60eWObUIW/ZZ6nFC2lzW6MmbmdAuKRLeEUSEDvMThXpx5ZxepIapWoBpp0RlTUhoUiTiTGSIweMTdFsFh1RH1uv16WxZTH6+fPneXhApFu3bpVU69IcZ1u8Si41Nckw6EsRV+c5wzYzMFfmNRwOt7a22CNsCu9fW1uT5wqraTBNTU3dvHmTMJuCVQ5PyneyKGEX3Tt2KsYaaJAB0llFxBniGnkdothpkcjYb7SnIGMWy5ej7XsPrE+PCROgeqx90UFyPbl4hSCRoP57gSpQiHNV0vQekqT6AwzB6Pf7HKHNBjllSmywdAyxCMFME+XQkoW9ZnggZ5kNrHan+GasuJScKTfKlg/r9XoGBezv729vbwtrzOhhxWJJyexFBMaWixEodcylXCE5ZIxAtpZ42XNhZ43YhgmU87FYRvcnEw7//ecsggzdFi/7LZFv9JiCOEvvXBtrkGnxZUeutQQglcG7DEeSvQJqWaanaYGSb+sSokWHJPs5X+T8JhIPAalWqw0GA7dUnME383kvX778DzDbxuU2ToEFAAAAAElFTkSuQmCC\"\n",
    "\n",
    "def load_test_image(size=(256, 256), name=None):\n",
    "    \"\"\"Return a grayscale float32 image in [0, 1].\"\"\"\n",
    "    if HAS_SKIMAGE:\n",
    "        if name is None or name.lower() == \"punch\":\n",
    "            img_data = base64.b64decode(_B64_IMG)\n",
    "            img_pil = Image.open(io.BytesIO(img_data)).convert(\"L\")\n",
    "            img = np.array(img_pil, dtype=np.float32) / 255.0\n",
    "        elif name.lower() == \"camera\":\n",
    "            img = skdata.camera().astype(np.float32) / 255.0\n",
    "            img = sk_resize(img, size, anti_aliasing=True).astype(np.float32)\n",
    "    else:\n",
    "        y, x = np.ogrid[:size[0], :size[1]]\n",
    "        cy, cx = size[0] / 2, size[1] / 2\n",
    "        r = np.sqrt((x - cx)**2 + (y - cy)**2)\n",
    "        img = (0.5 + 0.5 * np.sin(r * 0.4)).astype(np.float32)\n",
    "    return img.astype(np.float32)\n",
    "\n",
    "IMG = load_test_image()\n",
    "print(f\"Test image: shape={IMG.shape}, dtype={IMG.dtype}, range=[{IMG.min():.3f}, {IMG.max():.3f}]\")\n",
    "plt.imshow(IMG, cmap=\"gray\", vmin=0, vmax=1)\n",
    "plt.title(\"Test image\"); plt.axis(\"off\"); plt.show()\n",
    "\n",
    "# ── Utility: side-by-side image + histogram comparison ───────────────────────\n",
    "def show_histogram_comparison(img_orig, img_mod, mod_title=\"Modified\"):\n",
    "    \"\"\"Display original vs modified image with their normalised histograms.\"\"\"\n",
    "    fig, axes = plt.subplots(2, 2, figsize=(10, 6))\n",
    "    for ax, im, title in zip(axes[0], [img_orig, img_mod], [\"Original\", mod_title]):\n",
    "        ax.imshow(im, cmap=\"gray\", vmin=0, vmax=1)\n",
    "        ax.set_title(title); ax.axis(\"off\")\n",
    "    for ax, im, title in zip(axes[1], [img_orig, img_mod], [\"Original hist\", f\"{mod_title} hist\"]):\n",
    "        counts, bins = np.histogram(im, bins=256, range=(0., 1.))\n",
    "        ax.bar(bins[:-1], counts / im.size, width=1/256, color=\"steelblue\")\n",
    "        ax.set_xlim(0, 1); ax.set_title(title)\n",
    "    plt.tight_layout(); plt.show()\n",
    "\n",
    "# ── Kernel utilities ─────────────────────────────────────────────────────────\n",
    "def box_kernel(size):\n",
    "    \"\"\"Uniform box kernel of given size.\"\"\"\n",
    "    return np.ones((size, size), dtype=np.float32) / (size * size)\n",
    "\n",
    "def gaussian_kernel_2d(size, sigma):\n",
    "    \"\"\"Normalised 2-D Gaussian kernel.\"\"\"\n",
    "    k = size // 2\n",
    "    y, x = np.ogrid[-k:k+1, -k:k+1]\n",
    "    h = np.exp(-(x**2 + y**2) / (2.0 * sigma**2))\n",
    "    h /= h.sum()\n",
    "    return h.astype(np.float32)\n",
    "\n",
    "def gaussian_kernel_1d(size, sigma):\n",
    "    \"\"\"Normalised 1-D Gaussian kernel.\"\"\"\n",
    "    k = size // 2\n",
    "    x = np.arange(-k, k + 1, dtype=np.float64)\n",
    "    h = np.exp(-x**2 / (2.0 * sigma**2))\n",
    "    return (h / h.sum()).astype(np.float32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code1",
   "metadata": {},
   "outputs": [],
   "source": [
    "DPI = 100\n",
    "# Plotting defaults\n",
    "plt.rcParams['figure.figsize'] = (12, 5)\n",
    "plt.rcParams['image.cmap'] = 'gray'\n",
    "plt.rcParams['axes.titlesize'] = 11\n",
    "\n",
    "# ── Reusable save helper ──────────────────────────────────────────────\n",
    "def save(fig, slide_num):\n",
    "    \"\"\"Save figure as img/slideN.png at 100 dpi and display.\"\"\"\n",
    "    path = f'img/slide{slide_num}.png'\n",
    "    fig.savefig(path, dpi=DPI, bbox_inches='tight', facecolor='white')\n",
    "    plt.show()\n",
    "    plt.close(fig)\n",
    "    print(f'  ✓ saved {path}')\n",
    "\n",
    "# ── Test image ────────────────────────────────────────────────────────\n",
    "print(f\"Test image | shape: {IMG.shape}  dtype: {IMG.dtype}  \"\n",
    "      f\"range: [{IMG.min():.0f}, {IMG.max():.0f}]\")\n",
    "print(\"Setup complete.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55db5a5f",
   "metadata": {},
   "source": [
    "---\n",
    "# Warm-Up Review\n",
    "\n",
    "**Recall from last week:**\n",
    "- Linear filtering (convolution / correlation)\n",
    "- Box / Gaussian filters – smoothing to avoid aliasing\n",
    "- Correlation as \"matching\" (dot product → NCC)\n",
    "- Pyramid representations for scale variations"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8133be9",
   "metadata": {},
   "source": [
    "## Convolution Refresher\n",
    "*(generates the figure shown on the slide)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.ndimage import convolve\n",
    "\n",
    "# Demonstrate convolution step-by-step on a small patch\n",
    "patch = np.array([\n",
    "    [10, 10, 10, 80, 80],\n",
    "    [10, 10, 10, 80, 80],\n",
    "    [10, 10, 10, 80, 80],\n",
    "    [10, 10, 10, 80, 80],\n",
    "    [10, 10, 10, 80, 80],\n",
    "], dtype=float)\n",
    "\n",
    "kernel_avg = np.ones((3, 3)) / 9.0          # averaging (smoothing)\n",
    "kernel_dx  = np.array([[-1, 0, 1],\n",
    "                       [-1, 0, 1],\n",
    "                       [-1, 0, 1]], dtype=float)  # horizontal derivative\n",
    "\n",
    "out_smooth = convolve(patch, kernel_avg)\n",
    "out_deriv  = convolve(patch, kernel_dx)\n",
    "\n",
    "# Define the grid layout visually: '.' means an empty space\n",
    "layout = [\n",
    "    ['a', 'b', 'd'],\n",
    "    ['.', 'c', 'e']\n",
    "]\n",
    "\n",
    "# Create the figure using subplot_mosaic\n",
    "fig, axes = plt.subplot_mosaic(layout, figsize=(9, 6))\n",
    "\n",
    "# Map the mosaic keys to their respective data and titles\n",
    "plot_data = {\n",
    "    'a': (patch, 'Input patch'),\n",
    "    'b': (kernel_avg, 'Box kernel'),\n",
    "    'c': (out_smooth, ''),\n",
    "    'd': (kernel_dx, 'Horizontal derivative'),\n",
    "    'e': (out_deriv, '')\n",
    "}\n",
    "\n",
    "# Iterate through the dictionary to plot\n",
    "for key, (arr, title) in plot_data.items():\n",
    "    ax = axes[key]\n",
    "    if key in ['b','d']:\n",
    "        im = ax.imshow(arr)\n",
    "    else:\n",
    "        im = ax.imshow(arr, cmap='cividis')\n",
    "    ax.set_title(title, fontsize=10)\n",
    "    \n",
    "    # Add text annotations\n",
    "    for (j, i), val in np.ndenumerate(arr):\n",
    "        ax.text(i, j, f'{val:.1f}', ha='center', va='center',\n",
    "                fontsize=10, color='white' if val <= 10 and val != 0 and val!=1 else 'black')\n",
    "    \n",
    "    ax.set_xticks([])\n",
    "    ax.set_yticks([])\n",
    "\n",
    "fig.suptitle('Convolution: smoothing vs. derivative kernel', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1537e3fb",
   "metadata": {},
   "source": [
    "## The Test Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "42aa1d01-5207-4b63-a34b-84e95f2f19ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig,ax = plt.subplots(figsize=(5, 5))\n",
    "ax.imshow(IMG,cmap='gray')\n",
    "d = IMG.shape[0]\n",
    "ax.set_title('Johnny Silvercat (%dx%d)'%(d,d),fontsize=18)\n",
    "ax.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8258c8be",
   "metadata": {},
   "source": [
    "## What Is a Derivative of an Image?\n",
    "\n",
    "The derivative measures *how fast* a function changes.\n",
    "For a discrete row of pixels: $\\frac{dI}{dx}\\big|_i \\approx I[i+1] - I[i]$\n",
    "\n",
    "**Edge detection = finding where the intensity changes sharply.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.gridspec as gridspec\n",
    "from matplotlib.patches import Rectangle\n",
    "\n",
    "# Extract the row and compute the derivative\n",
    "row = IMG[250, :]  # a single row\n",
    "deriv = np.diff(row)\n",
    "\n",
    "# Create a figure with a GridSpec layout: 2 rows, 2 columns\n",
    "# The width_ratios makes the 1D plots a bit wider than the image\n",
    "fig = plt.figure(figsize=(10, 5))\n",
    "gs = gridspec.GridSpec(2, 2, width_ratios=[1, 1])\n",
    "\n",
    "# Assign axes to the grid\n",
    "ax_img = fig.add_subplot(gs[:, 0])           # Spans both rows in column 0\n",
    "ax_prof = fig.add_subplot(gs[0, 1])          # Top right\n",
    "ax_der = fig.add_subplot(gs[1, 1], sharex=ax_prof) # Bottom right (sharing x-axis)\n",
    "\n",
    "# --- Left: Plot the Image ---\n",
    "ax_img.imshow(IMG, cmap='gray')\n",
    "ax_img.set_title('Original Image')\n",
    "ax_img.axis('off')\n",
    "\n",
    "# Add a red rectangle to highlight row 250\n",
    "# xy=(x, y) of the top-left corner. We start at x=0, y=248 to give it a thickness of 4 pixels\n",
    "img_width = IMG.shape[1]\n",
    "rect = Rectangle((0, 248), width=img_width, height=4, \n",
    "                 linewidth=1.5, edgecolor='red', facecolor='none')\n",
    "ax_img.add_patch(rect)\n",
    "\n",
    "# --- Right Top: 1-D intensity profile ---\n",
    "ax_prof.plot(row, 'b-', lw=1.2)\n",
    "ax_prof.set_ylabel('Intensity I(x)')\n",
    "ax_prof.set_title('1-D intensity profile (row 250)')\n",
    "ax_prof.fill_between(range(len(row)), row, alpha=0.15)\n",
    "# Hide x-axis tick labels for the top plot since they share the bottom axis\n",
    "ax_prof.tick_params(labelbottom=False) \n",
    "\n",
    "# --- Right Bottom: Derivative ---\n",
    "ax_der.plot(deriv, 'r-', lw=1.2)\n",
    "ax_der.set_ylabel('dI/dx')\n",
    "ax_der.set_xlabel('pixel x')\n",
    "ax_der.set_title('Finite difference: peaks = edges')\n",
    "ax_der.axhline(0, color='gray', ls='--', lw=0.7)\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "115f052e",
   "metadata": {},
   "source": [
    "---\n",
    "# Block 1 — Gradient-Based Edge Detection\n",
    "\n",
    "Edges mark **boundaries** — object boundaries, surface creases, texture borders, shadows.\n",
    "We find them by computing the **image gradient**:\n",
    "\n",
    "$$\\nabla I = \\begin{pmatrix} \\partial I/\\partial x \\\\ \\partial I/\\partial y \\end{pmatrix}$$\n",
    "\n",
    "- **Magnitude** $|\\nabla I| = \\sqrt{I_x^2 + I_y^2}$ → edge strength\n",
    "- **Direction** $\\theta = \\arctan(I_y / I_x)$ → edge orientation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b35e8e25",
   "metadata": {},
   "source": [
    "## Sobel Gradients $G_x$ and $G_y$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code6",
   "metadata": {},
   "outputs": [],
   "source": [
    "sobel_x = np.array([[-1,0,1],[-2,0,2],[-1,0,1]], dtype=float)\n",
    "sobel_y = sobel_x.T\n",
    "\n",
    "Gx = convolve(IMG, sobel_x)\n",
    "Gy = convolve(IMG, sobel_y)\n",
    "mag = np.sqrt(Gx**2 + Gy**2)\n",
    "direction = np.arctan2(Gy, Gx)\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(8, 4))\n",
    "axes = axes.flatten()\n",
    "axes[0].imshow(Gx, cmap='RdBu_r'); axes[0].set_title('Gx (horizontal)',fontsize=16)\n",
    "axes[1].imshow(Gy, cmap='RdBu_r'); axes[1].set_title('Gy (vertical)',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(8, 4))\n",
    "axes = axes.flatten()\n",
    "axes[0].imshow(direction, cmap='hsv'); axes[0].set_title('Direction $\\Theta$',fontsize=16)\n",
    "axes[1].imshow(mag, cmap='hot'); axes[1].set_title('Gradient magnitude |∇I|',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ebe2263",
   "metadata": {},
   "source": [
    "## Comparing Roberts, Prewitt, and Sobel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code7",
   "metadata": {},
   "outputs": [],
   "source": [
    "roberts_x = np.array([[1, 0], [0, -1]], dtype=float)\n",
    "roberts_y = np.array([[0, 1], [-1, 0]], dtype=float)\n",
    "prewitt_x = np.array([[-1,0,1],[-1,0,1],[-1,0,1]], dtype=float)\n",
    "prewitt_y = prewitt_x.T\n",
    "\n",
    "def edge_mag_dir(img, kx, ky):\n",
    "    gx = convolve(img, kx)\n",
    "    gy = convolve(img, ky)\n",
    "    return np.sqrt(gx**2 + gy**2),np.arctan2(gy, gx)\n",
    "\n",
    "mag_roberts,dir_roberts = edge_mag_dir(IMG, roberts_x, roberts_y)\n",
    "mag_prewitt,dir_prewitt = edge_mag_dir(IMG, prewitt_x, prewitt_y)\n",
    "mag_sobel,dir_sobel   = edge_mag_dir(IMG, sobel_x, sobel_y)\n",
    "\n",
    "fig, axes = plt.subplots(2, 3, figsize=(12, 8))\n",
    "axes = axes.flatten()\n",
    "# axes[0].imshow(IMG, cmap='gray'); axes[0].set_title('Original')\n",
    "axes[0].imshow(mag_roberts, cmap='hot'); axes[0].set_title('Roberts',fontsize=18)\n",
    "axes[1].imshow(mag_prewitt, cmap='hot'); axes[1].set_title('Prewitt',fontsize=18)\n",
    "axes[2].imshow(mag_sobel, cmap='hot');   axes[2].set_title('Sobel',fontsize=18)\n",
    "axes[3].imshow(dir_roberts, cmap='hsv')\n",
    "axes[4].imshow(dir_prewitt, cmap='hsv')\n",
    "axes[5].imshow(dir_sobel, cmap='hsv')\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('Comparison of gradient operators', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd502445",
   "metadata": {},
   "source": [
    "## Zoomed Comparison on a Crop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d5a26fbf-6c7b-4590-b1e9-c038aa7d506e",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMG2 = IMG[200:400,:200]\n",
    "\n",
    "roberts_x = np.array([[1, 0], [0, -1]], dtype=float)\n",
    "roberts_y = np.array([[0, 1], [-1, 0]], dtype=float)\n",
    "prewitt_x = np.array([[-1,0,1],[-1,0,1],[-1,0,1]], dtype=float)\n",
    "prewitt_y = prewitt_x.T\n",
    "\n",
    "def edge_mag_dir(img, kx, ky):\n",
    "    gx = convolve(img, kx)\n",
    "    gy = convolve(img, ky)\n",
    "    return np.sqrt(gx**2 + gy**2),np.arctan2(gy, gx)\n",
    "\n",
    "mag_roberts,dir_roberts = edge_mag_dir(IMG2, roberts_x, roberts_y)\n",
    "mag_prewitt,dir_prewitt = edge_mag_dir(IMG2, prewitt_x, prewitt_y)\n",
    "mag_sobel,dir_sobel   = edge_mag_dir(IMG2, sobel_x, sobel_y)\n",
    "\n",
    "fig, axes = plt.subplots(2, 3, figsize=(12, 8))\n",
    "axes = axes.flatten()\n",
    "# axes[0].imshow(IMG, cmap='gray'); axes[0].set_title('Original')\n",
    "axes[0].imshow(mag_roberts, cmap='hot'); axes[0].set_title('Roberts',fontsize=18)\n",
    "axes[1].imshow(mag_prewitt, cmap='hot'); axes[1].set_title('Prewitt',fontsize=18)\n",
    "axes[2].imshow(mag_sobel, cmap='hot');   axes[2].set_title('Sobel',fontsize=18)\n",
    "axes[3].imshow(dir_roberts, cmap='hsv')\n",
    "axes[4].imshow(dir_prewitt, cmap='hsv')\n",
    "axes[5].imshow(dir_sobel, cmap='hsv')\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('Comparison of gradient operators', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b8b889e",
   "metadata": {},
   "source": [
    "## Problem: Noise Ruins Simple Derivatives\n",
    "\n",
    "Derivatives amplify high-frequency content **including noise**.\n",
    "Solution: smooth first with a Gaussian, *then* differentiate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code10",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(42)\n",
    "noisy = IMG + np.random.normal(0, 25, IMG.shape)\n",
    "\n",
    "mag_noisy,dir_noisy   = edge_mag_dir(noisy, sobel_x, sobel_y)\n",
    "smoothed    = gaussian_filter(noisy, sigma=2)\n",
    "mag_smooth,dir_smooth  = edge_mag_dir(smoothed, sobel_x, sobel_y)\n",
    "\n",
    "fig, axes = plt.subplots(1, 4, figsize=(10, 3))\n",
    "axes[0].imshow(noisy, cmap='gray');        axes[0].set_title('Noisy image')\n",
    "axes[1].imshow(mag_noisy, cmap='hot');     axes[1].set_title('Sobel on noisy (garbage!)')\n",
    "axes[2].imshow(smoothed, cmap='gray');     axes[2].set_title('Gaussian smoothed (σ=2)')\n",
    "axes[3].imshow(mag_smooth, cmap='hot');    axes[3].set_title('Sobel on smoothed (clean)')\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('Derivatives amplify noise → smooth first!', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6991101",
   "metadata": {},
   "source": [
    "## Derivative-of-Gaussian (DoG) Kernels\n",
    "\n",
    "Because convolution is **associative** we can merge smoothing and\n",
    "differentiation into a *single* kernel:\n",
    "\n",
    "$$\\frac{\\partial}{\\partial x}(G_\\sigma * I) = \\frac{\\partial G_\\sigma}{\\partial x} * I$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code11",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dog_kernel_x(sigma, half=None):\n",
    "    if half is None:\n",
    "        half = int(3 * sigma)\n",
    "    ax = np.arange(-half, half + 1, dtype=float)\n",
    "    X, Y = np.meshgrid(ax, ax)\n",
    "    G = np.exp(-(X**2 + Y**2) / (2 * sigma**2))\n",
    "    dGdx = -(X / sigma**2) * G\n",
    "    return dGdx - dGdx.mean()\n",
    "    # return dGdx\n",
    "\n",
    "sigmas_dog = [.5, 1, 2]\n",
    "fig, axes = plt.subplots(2, len(sigmas_dog), figsize=(12, 8))\n",
    "for col, s in enumerate(sigmas_dog):\n",
    "    k = dog_kernel_x(s)\n",
    "    axes[0, col].imshow(k, cmap='RdBu_r')\n",
    "    axes[0, col].set_title(f'DoG kernel σ={s:.1f}',fontsize=18)\n",
    "    axes[0, col].axis('off')\n",
    "\n",
    "    dog_out = convolve(IMG, k)\n",
    "    axes[1, col].imshow(np.abs(dog_out), cmap='hot')\n",
    "    axes[1, col].set_title(f'|DoG * I| σ={s}',fontsize=18)\n",
    "    axes[1, col].axis('off')\n",
    "\n",
    "# fig.suptitle('Derivative-of-Gaussian kernels at different scales', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "406f9318",
   "metadata": {},
   "source": [
    "## The Canny Edge Detector — Pipeline Stages\n",
    "\n",
    "1. Gaussian smoothing → suppress noise\n",
    "2. Gradient magnitude & direction → find edges\n",
    "3. Non-Maximum Suppression → thin to 1 px\n",
    "4. Hysteresis thresholding → keep strong, link weak"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code13",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show each stage of Canny manually\n",
    "# img_norm = IMG / 255.0\n",
    "\n",
    "# Stage 1: Gaussian smoothing\n",
    "stage1 = gaussian_filter(IMG, sigma=1.5)\n",
    "\n",
    "# Stage 2: Gradient magnitude + direction\n",
    "gx_c = convolve(stage1, sobel_x / 255.0)\n",
    "gy_c = convolve(stage1, sobel_y / 255.0)\n",
    "mag_c = np.sqrt(gx_c**2 + gy_c**2)\n",
    "\n",
    "# Stage 3: NMS (quantised, simplified)\n",
    "def simple_nms(mag, angle):\n",
    "    rows, cols = mag.shape\n",
    "    nms = np.zeros_like(mag)\n",
    "    angle_deg = np.rad2deg(angle) % 180\n",
    "    for i in range(1, rows - 1):\n",
    "        for j in range(1, cols - 1):\n",
    "            a = angle_deg[i, j]\n",
    "            if (0 <= a < 22.5) or (157.5 <= a < 180):\n",
    "                n1, n2 = mag[i, j-1], mag[i, j+1]\n",
    "            elif 22.5 <= a < 67.5:\n",
    "                n1, n2 = mag[i-1, j+1], mag[i+1, j-1]\n",
    "            elif 67.5 <= a < 112.5:\n",
    "                n1, n2 = mag[i-1, j], mag[i+1, j]\n",
    "            else:\n",
    "                n1, n2 = mag[i-1, j-1], mag[i+1, j+1]\n",
    "            nms[i, j] = mag[i, j] if mag[i, j] >= max(n1, n2) else 0\n",
    "    return nms\n",
    "\n",
    "dir_c = np.arctan2(gy_c, gx_c)\n",
    "# NMS on a smaller crop for speed\n",
    "crop = slice(100, 300), slice(100, 400)\n",
    "nms_result = simple_nms(mag_c[crop], dir_c[crop])\n",
    "\n",
    "# Stage 4: Full Canny\n",
    "stage4 = canny(IMG, sigma=1.5, low_threshold=0.2, high_threshold=0.4)\n",
    "\n",
    "fig, axes = plt.subplots(1, 4, figsize=(12, 3))\n",
    "axes[0].imshow(stage1, cmap='gray');              axes[0].set_title('1. Gaussian smoothing',fontsize=16)\n",
    "axes[1].imshow(mag_c, cmap='hot');                axes[1].set_title('2. Gradient magnitude',fontsize=16)\n",
    "axes[2].imshow(nms_result, cmap='bwr');          axes[2].set_title('3. After NMS (crop)',fontsize=16)\n",
    "axes[3].imshow(stage4, cmap='bwr');              axes[3].set_title('4. Final Canny output',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('Canny pipeline stages', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccb65e7e",
   "metadata": {},
   "source": [
    "## Non-Maximum Suppression (NMS) — Before & After"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code14",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show before and after NMS on a zoomed region\n",
    "crop2 = slice(200, 300), slice(180, 300)\n",
    "mag_crop = mag_c[crop2]\n",
    "nms_crop = simple_nms(mag_c[crop2], dir_c[crop2])\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(14, 4))\n",
    "axes[0].imshow(IMG[crop2], cmap='gray');  axes[0].set_title('Original (crop)',fontsize=16)\n",
    "axes[1].imshow(mag_crop, cmap='hot');     axes[1].set_title('Gradient magnitude (thick ridges)',fontsize=16)\n",
    "axes[2].imshow(nms_crop, cmap='bwr');     axes[2].set_title('After NMS (1-pixel wide)',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('NMS thins gradient ridges to single-pixel edges', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e3a05cc",
   "metadata": {},
   "source": [
    "## Hysteresis Thresholding\n",
    "\n",
    "Two thresholds $k_{\\text{lo}} < k_{\\text{hi}}$:\n",
    "- $\\geq k_{\\text{hi}}$: **strong** (always kept)\n",
    "- $\\in [k_{\\text{lo}}, k_{\\text{hi}})$: **weak** (kept only if connected to strong)\n",
    "- $< k_{\\text{lo}}$: suppressed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code15",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Demonstrate strong / weak / suppressed\n",
    "mag_full,dir_full = edge_mag_dir(gaussian_filter(IMG, 1.5), sobel_x, sobel_y)\n",
    "nms = simple_nms(mag_full, dir_full)\n",
    "lo, hi = 0.2, 0.4\n",
    "\n",
    "strong = nms >= hi\n",
    "weak   = (nms >= lo) & (nms < hi)\n",
    "suppressed = nms < lo\n",
    "\n",
    "# Color-code: green=strong, yellow=weak, black=suppressed\n",
    "vis = np.zeros((*IMG.shape, 3))\n",
    "vis[strong] = [0, 0.8, 0]       # green\n",
    "vis[weak]   = [0.9, 0.7, 0]     # yellow\n",
    "# suppressed stays black\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(10, 4.5))\n",
    "axes[0].imshow(nms, cmap='bwr'); axes[0].set_title('Non-maximum suppression',fontsize=16)\n",
    "axes[1].imshow(vis);                  axes[1].set_title('Green=strong; Yellow=weak;\\n Black=suppressed',fontsize=16)\n",
    "axes[2].imshow(canny(IMG, sigma=1.5, low_threshold=lo, high_threshold=hi),\n",
    "               cmap='bwr');          axes[2].set_title('Final Canny (after linking)',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.suptitle(f'Hysteresis: $k_{{lo}}$={lo}, $k_{{hi}}$={hi}', fontweight='bold',fontsize=18)\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "deeb4bcb",
   "metadata": {},
   "source": [
    "## Laplacian-of-Gaussian (LoG)\n",
    "\n",
    "An alternative: instead of peaks in the 1st derivative, look for\n",
    "**zero crossings** in the 2nd derivative.\n",
    "\n",
    "$$\\nabla^2 I = \\frac{\\partial^2 I}{\\partial x^2} + \\frac{\\partial^2 I}{\\partial y^2}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "93f451a8-d223-41ab-8d6a-c03005cd34b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def log_kernel(sigma, half=None):\n",
    "    if half is None:\n",
    "        half = int(4 * sigma)\n",
    "    ax = np.arange(-half, half + 1, dtype=float)\n",
    "    X, Y = np.meshgrid(ax, ax)\n",
    "    r2 = X**2 + Y**2\n",
    "    s2 = sigma**2\n",
    "    k = -(1 / (np.pi * s2**2)) * (1 - r2 / (2 * s2)) * np.exp(-r2 / (2 * s2))\n",
    "    return k - k.mean()\n",
    "\n",
    "# Show LoG kernel and zero crossings\n",
    "log_k = log_kernel(2,half=12)\n",
    "log_out = convolve(IMG, log_k)\n",
    "zero_cross = np.zeros_like(log_out, dtype=bool)\n",
    "for i in range(1, log_out.shape[0]-1):\n",
    "    for j in range(1, log_out.shape[1]-1):\n",
    "        patch = log_out[i-1:i+2, j-1:j+2]\n",
    "        if patch.min() < 0 < patch.max():\n",
    "            zero_cross[i, j] = True\n",
    "\n",
    "fig, axes = plt.subplots(2, 2, figsize=(8, 8))\n",
    "axes = axes.flatten()\n",
    "axes[0].imshow(log_k, cmap='RdBu_r'); axes[0].set_title('LoG kernel (σ=2,k=25)\\n\"Mexican hat\"',fontsize=16)\n",
    "axes[1].imshow(log_out, cmap='hot'); axes[1].set_title('LoG * I',fontsize=16)\n",
    "axes[2].imshow(zero_cross, cmap='bwr'); axes[2].set_title('Zero crossings',fontsize=16)\n",
    "axes[3].imshow(canny(IMG, sigma=2,low_threshold=.08, high_threshold=.16), cmap='bwr'); axes[3].set_title('Canny (σ=2,lo=0.08,hi=0.16)',fontsize=16)\n",
    "for ax in axes: ax.axis('off')\n",
    "# fig.suptitle('LoG edge detection via zero crossings', fontweight='bold')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da4eb491",
   "metadata": {},
   "source": [
    "## ✏️ Exercise 1 — Canny Pipeline Exploration (~10 min)\n",
    "\n",
    "1. Apply Canny to an image of your choice with default parameters.\n",
    "2. Sweep $\\sigma \\in \\{0.5, 1, 2, 3\\}$ and $k_{\\text{hi}} \\in \\{0.05, 0.10, 0.20\\}$;\n",
    "   display the resulting edge maps in a $4 \\times 3$ grid."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d71f182d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ✏️ Exercise 1 — fill in the code below\n",
    "\n",
    "sigmas_sweep = [0.5, 1, 2, 3]\n",
    "hi_sweep = [0.05, 0.10, 0.20]\n",
    "\n",
    "fig, axes = plt.subplots(len(sigmas_sweep), len(hi_sweep),\n",
    "                          figsize=(12, 14))\n",
    "\n",
    "for r, s in enumerate(sigmas_sweep):\n",
    "    for c, h in enumerate(hi_sweep):\n",
    "        lo = h / 2\n",
    "        # TODO: compute Canny edge map with sigma=s, low_threshold=lo, high_threshold=h\n",
    "        # e = ...\n",
    "        # axes[r, c].imshow(e, cmap='gray')\n",
    "        axes[r, c].set_title(f'σ={s}, hi={h}', fontsize=9)\n",
    "        axes[r, c].axis('off')\n",
    "\n",
    "fig.suptitle('Exercise 1: Canny parameter sweep', fontweight='bold')\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82515dd1",
   "metadata": {},
   "source": [
    "---\n",
    "# Block 2 — Morphological Operations\n",
    "\n",
    "After edge detection (or thresholding) you typically get a **messy binary mask**:\n",
    "small noise blobs, tiny holes, touching objects.\n",
    "\n",
    "**Morphological operations** clean up binary images by probing them with a\n",
    "small shape called a **structuring element** (SE).\n",
    "\n",
    "| Operation | Effect |\n",
    "|-----------|--------|\n",
    "| **Dilation** ($\\oplus$) | Grow foreground |\n",
    "| **Erosion** ($\\ominus$) | Shrink foreground |\n",
    "| **Opening** | Remove small bright noise (erode → dilate) |\n",
    "| **Closing** | Fill small dark holes (dilate → erode) |"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17aae547",
   "metadata": {},
   "source": [
    "## Closing Example\n",
    "*(opening removes noise outside, closing fills holes inside)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code25",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Full exercise 3 solution\n",
    "np.random.seed(42)\n",
    "mask = np.zeros((200, 300), dtype=bool)\n",
    "rr, cc = draw_disk((80, 100), 40, shape=mask.shape)\n",
    "mask[rr, cc] = True\n",
    "rr, cc = draw_disk((120, 220), 30, shape=mask.shape)\n",
    "mask[rr, cc] = True\n",
    "\n",
    "noisy_mask = random_noise(mask.astype(float), mode='s&p', amount=0.05) > 0.5\n",
    "se_ex3 = diamond(2)\n",
    "step1 = morphology.dilation(noisy_mask, se_ex3).astype(float)\n",
    "step2 = morphology.erosion(step1, se_ex3).astype(float)\n",
    "\n",
    "\n",
    "fig, axes = plt.subplots(3, 1, figsize=(4, 8))\n",
    "axes[0].imshow(noisy_mask, cmap='gray'); axes[0].set_title('Noisy image',fontsize=20)\n",
    "axes[1].imshow(step1, cmap='gray');      axes[1].set_title('dilation',fontsize=20)\n",
    "axes[2].imshow(step2, cmap='gray');      axes[2].set_title('dilation+erosion',fontsize=20)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9311654",
   "metadata": {},
   "source": [
    "## Opening Example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "09ef80b3-daff-4b05-bec9-dcc6f4fba723",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Full exercise 3 solution\n",
    "np.random.seed(42)\n",
    "mask = np.zeros((200, 300), dtype=bool)\n",
    "rr, cc = draw_disk((80, 100), 40, shape=mask.shape)\n",
    "mask[rr, cc] = True\n",
    "rr, cc = draw_disk((120, 220), 30, shape=mask.shape)\n",
    "mask[rr, cc] = True\n",
    "\n",
    "noisy_mask = random_noise(mask.astype(float), mode='s&p', amount=0.05) > 0.5\n",
    "se_ex3 = diamond(2)\n",
    "step1 = morphology.erosion(noisy_mask, se_ex3).astype(float)\n",
    "step2 = morphology.dilation(step1, se_ex3).astype(float)\n",
    "\n",
    "\n",
    "fig, axes = plt.subplots(3, 1, figsize=(4, 8))\n",
    "axes[0].imshow(noisy_mask, cmap='gray'); axes[0].set_title('Noisy image',fontsize=20)\n",
    "axes[1].imshow(step1, cmap='gray');      axes[1].set_title('erosion',fontsize=20)\n",
    "axes[2].imshow(step2, cmap='gray');      axes[2].set_title('erosion+dilation',fontsize=20)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7a457e9",
   "metadata": {},
   "source": [
    "## Combining Opening + Closing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b10435c0-d679-4264-a889-d45561e5637f",
   "metadata": {},
   "outputs": [],
   "source": [
    "se_ex3 = diamond(2)\n",
    "stepA = morphology.opening(noisy_mask, se_ex3).astype(float)\n",
    "stepB = morphology.closing(noisy_mask, se_ex3).astype(float)\n",
    "stepA1 = morphology.closing(stepA, se_ex3).astype(float)\n",
    "stepB1 = morphology.opening(stepB, se_ex3).astype(float)\n",
    "\n",
    "fig, axes = plt.subplots(2, 2, figsize=(8, 8))\n",
    "axes = axes.flatten()\n",
    "axes[0].imshow(stepA, cmap='gray'); axes[0].set_title('opening',fontsize=20)\n",
    "axes[1].imshow(stepA1, cmap='gray');      axes[1].set_title('opening+\\nclosing',fontsize=20)\n",
    "axes[2].imshow(stepB, cmap='gray');      axes[2].set_title('closing',fontsize=20)\n",
    "axes[3].imshow(stepB1, cmap='gray');      axes[3].set_title('closing+\\nopening',fontsize=20)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59e426fe",
   "metadata": {},
   "source": [
    "## Morphological Gradient (Edge Detector)\n",
    "\n",
    "$$\\text{grad}(I) = (I \\oplus B) - (I \\ominus B)$$\n",
    "\n",
    "Dilated minus eroded = thick edges at boundaries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code24",
   "metadata": {},
   "outputs": [],
   "source": [
    "se_mg = disk(3)\n",
    "morph_grad = morphology.dilation(IMG, se_mg).astype(float) - \\\n",
    "             morphology.erosion(IMG, se_mg).astype(float)\n",
    "\n",
    "sobel_mag,sobel_dir = edge_mag_dir(IMG, sobel_x, sobel_y)\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(10, 4))\n",
    "axes[0].imshow(IMG, cmap='gray');        axes[0].set_title('Original',fontsize=20)\n",
    "axes[1].imshow(morph_grad, cmap='hot');  axes[1].set_title('dilation − erosion',fontsize=20)\n",
    "axes[2].imshow(sobel_mag, cmap='hot');   axes[2].set_title('Sobel mag on original',fontsize=20)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51bf1652",
   "metadata": {},
   "source": [
    "## ✏️ Exercise 2 — Morphological Clean-Up (~10 min)\n",
    "\n",
    "Apply morphological operations to clean the noisy mask below.\n",
    "\n",
    "**Hint:** try opening followed by closing (or vice-versa) with a diamond(2) or disk(3) SE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b8de0ce2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ✏️ Exercise 2 — fill in the code below\n",
    "\n",
    "from skimage.morphology import diamond, disk\n",
    "from skimage.util import random_noise\n",
    "\n",
    "###############################################################################\n",
    "###############################################################################\n",
    "# IMAGE CREATION\n",
    "# Create a noisy binary mask\n",
    "np.random.seed(42)\n",
    "mask = np.zeros((400, 400), dtype=bool)\n",
    "\n",
    "# scattered cells / blobs\n",
    "for _ in range(15):\n",
    "    cy, cx = np.random.randint(40, 360, size=2)\n",
    "    r = np.random.randint(12, 35)\n",
    "    rr, cc = draw_disk((cy, cx), r, shape=mask.shape)\n",
    "    mask[rr, cc] = True\n",
    "\n",
    "# a long curved \"vessel\"\n",
    "t = np.linspace(0, 2 * np.pi, 800)\n",
    "yy = (200 + 120 * np.sin(t)).astype(int)\n",
    "xx = (np.linspace(30, 370, len(t))).astype(int)\n",
    "for dy in range(-3, 4):\n",
    "    mask[np.clip(yy + dy, 0, 399), xx] = True\n",
    "\n",
    "# rectangular \"tissue boundary\"\n",
    "mask[280:320, 50:350] = True\n",
    "mask[280:395, 160:200] = True\n",
    "\n",
    "# punch holes inside objects\n",
    "for _ in range(30):\n",
    "    cy, cx = np.random.randint(0, 400, size=2)\n",
    "    if mask[cy, cx]:\n",
    "        rr, cc = draw_disk((cy, cx), np.random.randint(2, 6), shape=mask.shape)\n",
    "        mask[rr, cc] = False\n",
    "\n",
    "# add salt-and-pepper + small blob noise\n",
    "noisy_mask = random_noise(mask.astype(float), mode='s&p', amount=0.06) > 0.5\n",
    "for _ in range(40):\n",
    "    cy, cx = np.random.randint(0, 400, size=2)\n",
    "    rr, cc = draw_disk((cy, cx), np.random.randint(1, 4), shape=noisy_mask.shape)\n",
    "    noisy_mask[rr, cc] = not mask[cy, cx]\n",
    "\n",
    "###############################################################################\n",
    "###############################################################################\n",
    "# YOUR WORK STARTS HERE\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(15, 4))\n",
    "axes[0].imshow(noisy_mask, cmap='gray'); axes[0].set_title('Noisy mask')\n",
    "\n",
    "# TODO: apply morphological operations to clean the mask\n",
    "# se = ...\n",
    "# cleaned = ...\n",
    "\n",
    "# TODO: plot result image (just uncomment)\n",
    "# axes[1].imshow(cleaned, cmap='gray'); axes[1].set_title('After cleaning')\n",
    "\n",
    "axes[2].imshow(mask, cmap='gray'); axes[2].set_title('Ground truth')\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f357aa65",
   "metadata": {},
   "source": [
    "---\n",
    "# Block 3 — Corners & Lines\n",
    "\n",
    "## The Aperture Problem\n",
    "\n",
    "- **Flat region** → slide any direction → no change → *boring*\n",
    "- **Edge** → slide along → no change; slide across → change → *ambiguous*\n",
    "- **Corner** → slide any direction → change → *distinctive!*\n",
    "\n",
    "Corners are ideal anchor points for matching, tracking, and stitching."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9eb736b2",
   "metadata": {},
   "source": [
    "## Aperture Problem — Visual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "222b303d-7555-4c8c-b8af-d9e4fe314740",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Harris Corner Detector — Gradient Distribution Plots\n",
    "=====================================================\n",
    "Generates scatter plots of (Ix, Iy) gradient vectors for four\n",
    "synthetic patch types: Flat, Linear Edge, Corner, and Vertical Edge.\n",
    "Each plot includes the principal-component ellipse fitted to the\n",
    "gradient cloud, matching the style from Hebert / Collins slides.\n",
    "\n",
    "Usage: run all cells in a Jupyter notebook, or run as a script.\n",
    "\"\"\"\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.patches import Ellipse\n",
    "# CHANGED: Import correlate instead of convolve\n",
    "from scipy.ndimage import correlate, gaussian_filter\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# 1.  Helper: compute Ix, Iy with Sobel\n",
    "# ---------------------------------------------------------------------------\n",
    "def sobel_gradients(patch):\n",
    "    \"\"\"Return (Ix, Iy) arrays for a greyscale patch.\"\"\"\n",
    "    Gx = np.array([[-1, 0, 1],\n",
    "                   [-2, 0, 2],\n",
    "                   [-1, 0, 1]], dtype=float) / 8.0   # normalise\n",
    "    Gy = Gx.T\n",
    "    \n",
    "    # CHANGED: Use correlate to prevent the kernel from flipping.\n",
    "    # This ensures transitions from dark (0) to bright (1) yield positive gradients.\n",
    "    Ix = correlate(patch.astype(float), Gx)\n",
    "    Iy = correlate(patch.astype(float), Gy)\n",
    "    return Ix, Iy\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# 2.  Helper: PCA ellipse from a 2-column data matrix\n",
    "# ---------------------------------------------------------------------------\n",
    "def pca_ellipse(Ix_flat, Iy_flat, n_std=2.0):\n",
    "    \"\"\"Return (centre, width, height, angle_deg) for matplotlib Ellipse.\"\"\"\n",
    "    D = np.column_stack([Ix_flat, Iy_flat])\n",
    "    \n",
    "    # CHANGED: Harris uses the uncentered Structure Tensor (Second Moment Matrix).\n",
    "    # We use (D.T @ D) / N instead of np.cov, which centers the data.\n",
    "    M = (D.T @ D) / len(D) \n",
    "    eigvals, eigvecs = np.linalg.eigh(M)\n",
    "    \n",
    "    order = eigvals.argsort()[::-1]\n",
    "    eigvals = eigvals[order]\n",
    "    eigvecs = eigvecs[:, order]\n",
    "    \n",
    "    angle = np.degrees(np.arctan2(eigvecs[1, 0], eigvecs[0, 0]))\n",
    "    width  = 2 * n_std * np.sqrt(max(eigvals[0], 1e-10))\n",
    "    height = 2 * n_std * np.sqrt(max(eigvals[1], 1e-10))\n",
    "    \n",
    "    # CHANGED: Force center to origin, consistent with Structure Tensor math\n",
    "    return (0.0, 0.0), width, height, angle\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# 3.  Build four synthetic patches  (64 x 64, values in [0, 1])\n",
    "# ---------------------------------------------------------------------------\n",
    "np.random.seed(42)\n",
    "sz = 64\n",
    "noise_std = 0.015       # subtle noise — keeps flat region tight\n",
    "\n",
    "def add_noise(patch):\n",
    "    return patch + noise_std * np.random.randn(*patch.shape)\n",
    "\n",
    "# --- Flat: uniform grey + light noise\n",
    "flat_patch = add_noise(0.5 * np.ones((sz, sz)))\n",
    "\n",
    "# --- Horizontal edge (dark top, bright bottom) — \"Linear Edge\"\n",
    "hedge_base = np.zeros((sz, sz))\n",
    "hedge_base[sz // 2:, :] = 1.0\n",
    "hedge_base = gaussian_filter(hedge_base, sigma=1.2)\n",
    "hedge_patch = add_noise(hedge_base)\n",
    "\n",
    "# --- Corner (dark top-left quadrant, bright elsewhere)\n",
    "corner_base = np.ones((sz, sz))\n",
    "corner_base[: sz // 2, : sz // 2] = 0.0\n",
    "corner_base = gaussian_filter(corner_base, sigma=1.2)\n",
    "corner_patch = add_noise(corner_base)\n",
    "\n",
    "# --- Vertical edge (dark left, bright right)\n",
    "vedge_base = np.zeros((sz, sz))\n",
    "vedge_base[:, sz // 2:] = 1.0\n",
    "vedge_base = gaussian_filter(vedge_base, sigma=1.2)\n",
    "vedge_patch = add_noise(vedge_base)\n",
    "\n",
    "patches = {\n",
    "    \"Flat\":           flat_patch,\n",
    "    \"H. Edge\":    hedge_patch,\n",
    "    \"Corner\":         corner_patch,\n",
    "    \"V. Edge\":  vedge_patch,\n",
    "}\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# 4.  Eigenvalue annotations for each case\n",
    "# ---------------------------------------------------------------------------\n",
    "eigen_labels = {\n",
    "    \"Flat\":          r\"$\\lambda_1 \\approx \\lambda_2 \\approx$ small\",\n",
    "    \"H. Edge\":   r\"$\\lambda_1$ large; $\\lambda_2 \\approx$ small\",\n",
    "    \"Corner\":        r\"$\\lambda_1 \\approx \\lambda_2 \\approx$ large\",\n",
    "    \"V. Edge\": r\"$\\lambda_1$ large; $\\lambda_2 \\approx$ small\",\n",
    "}\n",
    "\n",
    "# ---------------------------------------------------------------------------\n",
    "# 5.  Plot\n",
    "# ---------------------------------------------------------------------------\n",
    "fig, axes = plt.subplots(2, 2, figsize=(8, 8))\n",
    "axes = axes.ravel()\n",
    "\n",
    "for idx, (name, patch) in enumerate(patches.items()):\n",
    "    ax = axes[idx]\n",
    "\n",
    "    # Gradients\n",
    "    Ix, Iy = sobel_gradients(patch)\n",
    "\n",
    "    # Crop 2-pixel border to avoid boundary artefacts\n",
    "    Ix_inner = Ix[2:-2, 2:-2].ravel()\n",
    "    Iy_inner = Iy[2:-2, 2:-2].ravel()\n",
    "\n",
    "    # Scatter\n",
    "    ax.scatter(Ix_inner, Iy_inner, s=5, alpha=0.55, color=\"royalblue\",\n",
    "               edgecolors=\"none\", zorder=2)\n",
    "\n",
    "    # PCA ellipse\n",
    "    mu, w, h, angle = pca_ellipse(Ix_inner, Iy_inner, n_std=2.5)\n",
    "    ell = Ellipse(xy=mu, width=w, height=h, angle=angle,\n",
    "                  edgecolor=\"red\", facecolor=\"none\", linewidth=2.0)\n",
    "    ax.add_patch(ell)\n",
    "\n",
    "    # Axes through origin — auto-scale per subplot\n",
    "    lim = max(0.15, 1.3 * max(abs(Ix_inner).max(), abs(Iy_inner).max()))\n",
    "    ax.axhline(0, color=\"k\", linewidth=1.0)\n",
    "    ax.axvline(0, color=\"k\", linewidth=1.0)\n",
    "    ax.set_xlim(-lim, lim)\n",
    "    ax.set_ylim(-lim, lim)\n",
    "    ax.set_aspect(\"equal\")\n",
    "    ax.set_xlabel(r\"$I_x$\", fontsize=14)\n",
    "    ax.set_ylabel(r\"$I_y$\", fontsize=14)\n",
    "    ax.tick_params(labelsize=9)\n",
    "\n",
    "    # Title + eigenvalue annotation\n",
    "    ax.set_title(name, fontsize=16, fontweight=\"bold\", pad=10)\n",
    "    ax.text(0.03, 0.03, eigen_labels[name], transform=ax.transAxes,\n",
    "            fontsize=12, verticalalignment=\"bottom\",\n",
    "            bbox=dict(boxstyle=\"round,pad=0.3\", fc=\"white\", ec=\"gray\",\n",
    "                      alpha=0.9))\n",
    "\n",
    "    # Inset: the original patch\n",
    "    inset = ax.inset_axes([0.6, 0.6, 0.38, 0.38])\n",
    "    inset.imshow(patch, cmap=\"gray\", vmin=0, vmax=1)\n",
    "    inset.set_xticks([])\n",
    "    inset.set_yticks([])\n",
    "    for spine in inset.spines.values():\n",
    "        spine.set_edgecolor(\"darkorange\")\n",
    "        spine.set_linewidth(2.5)\n",
    "\n",
    "plt.tight_layout(pad=2.0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "becdf26a-8d55-49c3-8d38-9b6517ae76cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, axes = plt.subplots(2, 2, figsize=(8, 8))\n",
    "axes = axes.flatten()\n",
    "for idx, (name, patch) in enumerate(patches.items()):\n",
    "    ax = axes[idx]\n",
    "    ax.set_title(name, fontsize=16, fontweight=\"bold\")\n",
    "    ax.imshow(patch)\n",
    "    D = patch.shape[0]    \n",
    "    rect = plt.Rectangle((0,0), D-1,D-1, fill=False, color=\"orange\", linewidth=3)    \n",
    "    ax.add_patch(rect)\n",
    "    ax.axis('off')\n",
    "\n",
    "# plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "506166f5",
   "metadata": {},
   "source": [
    "## Harris Corner Detector\n",
    "\n",
    "$$R = \\det(M) - k\\,(\\operatorname{tr} M)^2$$\n",
    "\n",
    "- $R \\gg 0$: **corner**\n",
    "- $R \\ll 0$: edge\n",
    "- $|R| \\approx 0$: flat region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "code29",
   "metadata": {},
   "outputs": [],
   "source": [
    "checker = data.checkerboard().astype(float)\n",
    "# 1. Compute the Harris response map\n",
    "R = corner_harris(checker, k=0.05, sigma=1.0)\n",
    "# 2. Find corner locations (peaks in R)\n",
    "corners = corner_peaks(R, min_distance=10,threshold_rel=0.02)\n",
    "\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(12, 4.5))\n",
    "axes[0].imshow(checker, cmap='gray'); axes[0].set_title('Original',fontsize=20)\n",
    "axes[1].imshow(R, cmap='RdBu_r'); axes[1].set_title('Harris R',fontsize=20)\n",
    "axes[2].imshow(checker, cmap='gray')\n",
    "axes[2].plot(corners[:, 1], corners[:, 0], 'r+', markersize=12, mew=1.5)\n",
    "axes[2].set_title(f'Detected corners ({len(corners)})',fontsize=20)\n",
    "for ax in axes: ax.axis('off')\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5554ab67",
   "metadata": {},
   "source": [
    "## Hough Transform for Lines\n",
    "\n",
    "Each edge pixel casts a \"vote\" for every candidate line through it.\n",
    "Peaks in the accumulator = strong lines.\n",
    "\n",
    "$$x\\cos\\theta + y\\sin\\theta = \\rho$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9bb7fcd5-972d-4686-9b65-bd7ea948d6db",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a clean synthetic line image\n",
    "np.random.seed(42)\n",
    "line_img = np.zeros((200, 300), dtype=np.uint8)\n",
    "rr, cc = draw_line_sk(20, 30, 180, 270)\n",
    "rr = np.clip(rr, 0, 199); cc = np.clip(cc, 0, 299)\n",
    "line_img[rr, cc] = 255\n",
    "rr2, cc2 = draw_line_sk(10, 250, 190, 50)\n",
    "rr2 = np.clip(rr2, 0, 199); cc2 = np.clip(cc2, 0, 299)\n",
    "line_img[rr2, cc2] = 255\n",
    "rr3, cc3 = draw_line_sk(140, 10, 140, 290)\n",
    "rr3 = np.clip(rr3, 0, 199); cc3 = np.clip(cc3, 0, 299)\n",
    "line_img[rr3, cc3] = 255\n",
    "\n",
    "# Add noise\n",
    "noisy_lines = line_img.astype(float) + np.random.normal(0, 20, line_img.shape)\n",
    "edge_lines = canny(noisy_lines / 255.0, sigma=1)\n",
    "\n",
    "tested_angles = np.linspace(-np.pi/2, np.pi/2, 360, endpoint=False)\n",
    "h_space, theta_arr, dist_arr = hough_line(edge_lines, theta=tested_angles)\n",
    "\n",
    "# 1. Stricter peak extraction to ignore noise\n",
    "# We increase the minimum thresholds and explicitly limit to our top 3 lines\n",
    "hspace_peaks, angles, dists = hough_line_peaks(\n",
    "    h_space, \n",
    "    theta_arr, \n",
    "    dist_arr, \n",
    "    min_distance=20, \n",
    "    min_angle=15,\n",
    "    threshold=0.4 * np.max(h_space), # Higher threshold filters out weaker noise lines\n",
    "    num_peaks=3                     # Forces it to only keep the top 3 strongest votes\n",
    ")\n",
    "\n",
    "# 2. Create the enhanced visualization\n",
    "fig, axes = plt.subplots(2,2,figsize=(10, 8))\n",
    "axes = axes.flatten()\n",
    "\n",
    "axes[0].imshow(noisy_lines, cmap='gray'); axes[0].set_title('Noisy lines',fontsize=18)\n",
    "######################################################################\n",
    "axes[1].imshow(edge_lines, cmap='gray'); axes[1].set_title('Canny',fontsize=18)\n",
    "######################################################################\n",
    "\n",
    "###########################################################\n",
    "axes[2].set_facecolor('black')\n",
    "axes[2].imshow(np.log1p(h_space), aspect='auto', cmap='viridis', alpha=1,\n",
    "          extent=[np.rad2deg(theta_arr[0]), np.rad2deg(theta_arr[-1]),\n",
    "                  dist_arr[-1], dist_arr[0]])\n",
    "# Overlay the detected peaks as sharp, distinct points\n",
    "angles_deg = np.rad2deg(angles)\n",
    "axes[2].plot(angles_deg, dists, 'co', markersize=8, markeredgecolor='white', \n",
    "        label=f'{len(angles)} Lines Detected')\n",
    "\n",
    "axes[2].set_xlabel('θ (degrees)')\n",
    "axes[2].set_ylabel('ρ (pixels)')\n",
    "axes[2].set_title('Hough accumulator space',fontsize=18)\n",
    "axes[2].legend()\n",
    "################################################\n",
    "axes[3].imshow(noisy_lines, cmap='gray'); axes[3].set_title('Found lines',fontsize=18)\n",
    "for angle, dist in zip(angles, dists):\n",
    "    (x0, y0) = dist * np.array([np.cos(angle), np.sin(angle)])\n",
    "    axes[3].axline((x0, y0), slope=np.tan(angle + np.pi/2),\n",
    "                   color='red', linewidth=1.5, alpha=0.8)\n",
    "\n",
    "axes[3].set_xlim(0, noisy_lines.shape[1])\n",
    "axes[3].set_ylim(noisy_lines.shape[0], 0)\n",
    "for i in [0,1,3]: axes[i].axis('off')\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b4b1c14",
   "metadata": {},
   "source": [
    "## ✏️ Exercise 3 — Corners, Hough & RANSAC (~5 min)\n",
    "\n",
    "1. Apply the Harris detector to Johnny Silvercat; overlay detected corners.\n",
    "2. Detect 15 lines with the Hough transform; overlay them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a85d70de",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ✏️ Exercise 3 — fill in the code below\n",
    "\n",
    "from skimage.feature import corner_harris, corner_peaks\n",
    "from skimage.transform import hough_line, hough_line_peaks\n",
    "\n",
    "R = corner_harris(IMG, k=0.05, sigma=1.0)\n",
    "# TODO: find corner locations using corner_peaks\n",
    "# corners = corner_peaks(R, ...)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(6, 6))\n",
    "ax.imshow(IMG, cmap='gray')\n",
    "# TODO: overlay corners (just uncomment)\n",
    "# ax.plot(corners[:, 1], corners[:, 0], 'r+', markersize=8)\n",
    "ax.set_title('Harris corners and \\nHough lines on Johnny Silvercat',fontsize=18)\n",
    "\n",
    "\n",
    "# Hough lines\n",
    "edge_lines = canny(IMG, sigma=1)\n",
    "tested_angles = np.linspace(-np.pi/2, np.pi/2, 360, endpoint=False)\n",
    "h_space, theta_arr, dist_arr = hough_line(edge_lines, theta=tested_angles)\n",
    "\n",
    "# TODO: find 15 line peaks with hough_line_peaks\n",
    "# hspace_peaks, angles, dists = hough_line_peaks( ...\n",
    "\n",
    "# TODO: overlay lines (just uncomment)\n",
    "# for angle, dist in zip(angles, dists):\n",
    "#     (x0, y0) = dist * np.array([np.cos(angle), np.sin(angle)])\n",
    "#     ax.axline((x0, y0), slope=np.tan(angle + np.pi/2),\n",
    "                   # color='blue', linewidth=1, alpha=0.5)\n",
    "\n",
    "ax.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3994d3b",
   "metadata": {},
   "source": [
    "---\n",
    "# Summary\n",
    "\n",
    "| 1. Edge Detection | 2. Morphology | 3. Corners | 4. Lines & Models |\n",
    "|---|---|---|---|\n",
    "| Gradients highlight intensity changes | Clean up binary masks with SEs | Harris: gradient strong in *multiple* directions | Vote in parameter space (Hough) or sample (RANSAC) |\n",
    "| Canny = smooth → gradient → NMS → hysteresis | Dilation / Erosion / Opening / Closing | $R = \\det(M) - k(\\operatorname{tr}M)^2$ | $x\\cos\\theta + y\\sin\\theta = \\rho$ |"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c42b9c40",
   "metadata": {},
   "source": [
    "## Learning Goals — Self-Check\n",
    "\n",
    "Can you now…\n",
    "\n",
    "1. ✅ **Explain** what an image gradient is and how it relates to edges.\n",
    "2. ✅ **Write** the Sobel kernels from memory and apply them to a patch.\n",
    "3. ✅ **Name** the four steps of the Canny pipeline and explain what each one does.\n",
    "4. ✅ **Distinguish** dilation, erosion, opening, and closing — and know when to use each.\n",
    "5. ✅ **Clean up** a noisy binary mask using morphological operations in code.\n",
    "6. ✅ **Explain** why corners are more distinctive than edges, and what the Harris response $R$ tells you.\n",
    "7. ✅ **Describe** how the Hough transform uses voting to detect lines."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}