feat(review): 取证层多源化 + 视觉多图（整体+2x2 象限）

- 新增 evidence_model.py：统一证据模型（source_type + images 列表），访问器兼容旧 image_b64/image_mime 单图字段
- phases/evidence.py：新增 collect() 多源分发 seam（仅 CAD）、capture_images() 整体+等分象限多图（结束复位视口）、sub_windows()
- phases/review.py：多图逐张下采样喂 LLM，payload 增加 source_type
- config.py：EVIDENCE_DETAIL_TILES=4 / EVIDENCE_MAX_IMAGES=5（默认开启多图）
- 新增 tests/test_evidence_model.py，更新 evidence/review 断言与 fakes 归一

dwg_review/phases/review.py

@@ -11,6 +11,7 @@
 
 import json
 
+from .. import evidence_model
 from ..cad_tools import READ_ONLY_TOOLS, is_write_tool
 from ..findings import parse_conclusion
 from ..llm_client import LlmClient, downsample_image_b64
@@ -24,6 +25,7 @@ def evidence_payload(evidence):
     """把证据整理成给 LLM 的 JSON（剔除 base64 图片，避免撑爆上下文）。"""
     frame = evidence.get("frame") or {}
     return {
+        "source_type": evidence.get("source_type"),
         "frame_index": evidence.get("index"),
         "frame": {
             k: frame.get(k)
@@ -72,9 +74,11 @@ def review_frame(ctx, evidence, system_prompt, log=print):
             + json.dumps(payload, ensure_ascii=False, indent=2),
         }
     ]
-    image_b64 = evidence.get("image_b64")
-    if image_b64:
-        nb, nm = downsample_image_b64(image_b64, evidence.get("image_mime") or "image/png")
+    for img in evidence_model.evidence_images(evidence):
+        nb, nm = downsample_image_b64(img["b64"], img.get("mime") or "image/png")
+        label = img.get("label")
+        if label:
+            user_content.append({"type": "text", "text": f"（截图：{label}）"})
         user_content.append(LlmClient.image_content(nb, nm))
 
     messages = [