feat: v3 robustness upgrade — EasyOCR, failure recovery, minimum content check

- OCR: EasyOCR (primary, ch_sim+en) with PaddleOCR fallback for Windows compatibility
- Validation: _check_minimum_content() rejects empty-shell JRXML (no band/textField)
- Retry: MAX_RETRY 3→5, exhaustion records pending_failure_context for next-turn auto-injection
- Finalize: only saves jrxml_versions on pass, preserves last good final_jrxml on fail
- Extract JRXML: improved empty markdown block handling and XML fragment fallback
- UI: real-time node progress via placeholder updates, initial "analyzing" feedback
- UI: use agent_state (full) instead of node_state (partial) for summary card routing
- UI: unknown template_type now gives LLM meaningful image context instead of metadata
- Docs: updated CLAUDE.md and CODE_GUIDE.md to reflect all v3 changes

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

backend/file_parser.py

@@ -65,17 +65,36 @@ def parse_file(file_path: str, file_type: str = "") -> dict:
 # ---------------------------------------------------------------------------
 
 def _parse_image(path: Path) -> dict:
-    """OCR 提取图片中的文字。"""
+    """OCR 提取图片中的文字。优先 EasyOCR，回退 PaddleOCR。"""
     try:
         img = PIL.Image.open(path)
         info = f"[图片: {img.size[0]}x{img.size[1]}, {img.mode}]"
     except Exception:
         info = "[图片: 无法读取元数据]"
 
-    # 尝试 PaddleOCR
+    # 优先 EasyOCR（Windows 兼容性更好）
+    try:
+        import easyocr
+        import numpy as np
+        reader = easyocr.Reader(["ch_sim", "en"], gpu=False, verbose=False)
+        result = reader.readtext(np.array(img))
+        lines = [text.strip() for (_, text, _) in result if text.strip()]
+        if lines:
+            return {
+                "text": f"{info}\n识别文本:\n" + "\n".join(lines),
+                "file_type": "image",
+                "method": "easyocr",
+                "error": None,
+            }
+    except ImportError:
+        pass
+    except Exception:
+        pass
+
+    # 回退 PaddleOCR
     try:
         from paddleocr import PaddleOCR
-        ocr = PaddleOCR(lang="ch", use_angle_cls=False, show_log=False)
+        ocr = PaddleOCR(lang="ch")
         result = ocr.ocr(str(path))
         lines = []
         if result and result[0]:
@@ -97,7 +116,7 @@ def _parse_image(path: Path) -> dict:
 
     # OCR 不可用 → 返回图片元信息 + 安装提示
     return {
-        "text": f"{info}\n(如需 OCR 文字识别，请安装: pip install paddleocr)",
+        "text": f"{info}\n(如需 OCR 文字识别，请安装: pip install easyocr)",
         "file_type": "image",
         "method": "metadata_only",
         "error": "OCR 引擎未安装，已返回图片元信息",