docs: 交接文档三件套(接手总览+后端地图+差距路线图)+补全env模板

- HANDOVER.md: 守印接手第一入口,全景+上手路径+文档地图
- backend/BACKEND_MAP.md: 后端接手地图(路由/模型/AI引擎/评分维度)
- 路线图.md: 现状vs目标差距(三视角×双目标矩阵+7/15倒排冲刺)
- .env.example x3: 补全 MODEL_VISION/CLAUDE_API_URL 及本地SSE示例

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.env.example

@@ -29,8 +29,12 @@ IMAGE_PROVIDER_FALLBACK=gemini
 IMAGE_MODEL=gpt-image-2
 MODEL_IMAGE=gpt-image-2
 MODEL_TEXT=claude-opus-4-8
+# Vision model used by verify_real_image script (leave empty to skip real-image checks).
+MODEL_VISION=
 APIPORTS_BASE_URL=https://www.apiports.com/v1
 CODEPROXY_BASE_URL=
+# Override only if calling Claude through a non-default gateway.
+CLAUDE_API_URL=
 GEMINI_API_URL=https://generativelanguage.googleapis.com/v1beta
 
 # Optional platform keys. Normal customer keys should be entered in the UI.