feat: v4 multimodal chat input, multi-format support, and annotation detection

- Replace st.chat_input with st-multimodal-chatinput (Ctrl+V paste, drag-drop, file button)
- Extract _process_uploaded_file() shared handler (eliminates ~70 duplicated lines)
- Add XLSX (openpyxl), XLS (xlrd), DOC (olefile) parsers to file_parser.py
- Add backend/annotation_detector.py: circle detection (HoughCircles) + arrow detection (HoughLinesP clustering) + OCR correlation + LLM context formatting
- Add annotation_result field to AgentState with session persistence
- Wire annotation detection into process_input and _format_ocr_context
- Add 11 new tests: 7 annotation detector + 4 multi-format parser
- Update all docs: CLAUDE.md, README.md, CODE_GUIDE.md, ROADMAP.md

requirements.txt

@@ -26,6 +26,24 @@ python-dotenv>=1.0.0
 httpx>=0.27.0
 tiktoken>=0.7.0
 
+# OCR 依赖（PaddleOCR 精确识别优先，EasyOCR 回退）
+# Pinned: paddleocr 2.9.x + paddlepaddle 2.6.x known-stable on Windows CPU
+# 3.x has ONEDNN compatibility issues on Windows
+paddleocr>=2.9.0,<3.0.0
+paddlepaddle>=2.6.0,<3.0.0
+easyocr>=1.7.0
+# 聊天输入增强（粘贴/拖拽上传）
+st-multimodal-chatinput>=0.2.1
+
+# 多格式文件解析
+openpyxl>=3.1.0
+xlrd>=2.0.0
+olefile>=0.47
+
+# 批注检测（圈选/箭头识别）
+opencv-python-headless>=4.8.0
+
 # 测试
 pytest>=8.0.0
 pytest-asyncio>=0.24.0
+xlwt>=1.3.0