feat: multimodal image clients, workspaces, Tushare skill, test fixes

- Replace monolithic image_message_client with HTTP/OCR/legacy modules; tighten OpenAI transport + tool schemas for multimodal downgrade to OCR specialist path.
- Add image/stock workspace prompts (META/SOUL/ROLE_SYSTEM); register experts; tweak specialist agent/direct loop/query_image_attachment.
- Add bundled runtime/skills/tushare-finance (references, api_client, SKILL metadata).
- Document OCR-related env vars; admin chat tweaks; README; weixin_install Ensure-OfficialPluginRuntimeDeps helper.
- Tests: multimodal downgrade + OCR coverage, strict tool pairing in attachment replay guard, workspace contract skips _internal/_system dirs, router/trace/prompt guards.

Co-authored-by: Cursor <cursoragent@cursor.com>

runtime/skills/tushare-finance/reference/接口文档/复权因子.md

@@ -0,0 +1,125 @@
+# 复权因子
+
+**文档ID**: 28
+**原始链接**: https://tushare.pro/document/2?doc_id=28
+
+---
+
+## 复权因子
+
+接口：adj_factor，可以通过数据工具调试和查看数据。更新时间：盘前9点15~20分完成当日复权因子入库描述：本接口由Tushare自行生产，获取股票复权因子，可提取单只股票全部历史复权因子，也可以提取单日全部股票的复权因子。积分要求：2000积分起，5000以上可高频调取
+
+输入参数
+
+<table>
+<thead>
+<tr>
+<th>名称</th>
+<th>类型</th>
+<th>必选</th>
+<th>描述</th>
+</tr>
+</thead>
+<tbody><tr>
+<td>ts_code</td>
+<td>str</td>
+<td>N</td>
+<td>股票代码</td>
+</tr>
+<tr>
+<td>trade_date</td>
+<td>str</td>
+<td>N</td>
+<td>交易日期(YYYYMMDD，下同)</td>
+</tr>
+<tr>
+<td>start_date</td>
+<td>str</td>
+<td>N</td>
+<td>开始日期</td>
+</tr>
+<tr>
+<td>end_date</td>
+<td>str</td>
+<td>N</td>
+<td>结束日期</td>
+</tr>
+</tbody></table>
+注：日期都填YYYYMMDD格式，比如20181010
+
+输出参数
+
+<table>
+<thead>
+<tr>
+<th>名称</th>
+<th>类型</th>
+<th>描述</th>
+</tr>
+</thead>
+<tbody><tr>
+<td>ts_code</td>
+<td>str</td>
+<td>股票代码</td>
+</tr>
+<tr>
+<td>trade_date</td>
+<td>str</td>
+<td>交易日期</td>
+</tr>
+<tr>
+<td>adj_factor</td>
+<td>float</td>
+<td>复权因子</td>
+</tr>
+</tbody></table>
+接口示例
+
+```
+
+pro = ts.pro_api()
+
+#提取000001全部复权因子
+df = pro.adj_factor(ts_code='000001.SZ', trade_date='')
+
+
+#提取2018年7月18日复权因子
+df = pro.adj_factor(ts_code='', trade_date='20180718')
+
+```
+
+或者
+
+```
+
+df = pro.query('adj_factor',  trade_date='20180718')
+
+```
+
+数据样例
+
+```
+        ts_code trade_date  adj_factor
+0     000001.SZ   20180809     108.031
+1     000001.SZ   20180808     108.031
+2     000001.SZ   20180807     108.031
+3     000001.SZ   20180806     108.031
+4     000001.SZ   20180803     108.031
+5     000001.SZ   20180802     108.031
+6     000001.SZ   20180801     108.031
+7     000001.SZ   20180731     108.031
+8     000001.SZ   20180730     108.031
+9     000001.SZ   20180727     108.031
+10    000001.SZ   20180726     108.031
+11    000001.SZ   20180725     108.031
+12    000001.SZ   20180724     108.031
+13    000001.SZ   20180723     108.031
+14    000001.SZ   20180720     108.031
+15    000001.SZ   20180719     108.031
+16    000001.SZ   20180718     108.031
+17    000001.SZ   20180717     108.031
+18    000001.SZ   20180716     108.031
+19    000001.SZ   20180713     108.031
+20    000001.SZ   20180712     108.031
+
+```