chore(prep): shared groundwork for batch-4 (report pipeline)

第四批（T15 编排/T16 视觉诊断/T17 文案校验）协调提交：
- 迁移 0006：rp_report（result JSONB 按阶段键承载中间态）/rp_input
 （derived_features 视觉中间态）/rp_share（V1 只建结构）
- ORM 三模型；错误码 40420/40920（文案无价格字样）
- 配置：report_pipeline_stages（dotted path 动态装配）+ vision/tier
- packages/report/pipeline.py：Stage Protocol + ReportContext（三任务共同契约，
  动态 import 的动机与代价写在 docstring）
- 路由桩 reports.py 已注册

packages/shared/config.py

@@ -213,6 +213,24 @@ class Settings(BaseSettings):
     retrieval_cache_ttl_seconds: int = 60
     retrieval_cache_max_entries: int = 512
 
+    # --- 美学报告流水线（T-P3，第四批）---------------------------------------
+    # 阶段按 dotted path 动态装配（importlib）：T15 编排、T16 供 vision/diagnosis、
+    # T17 供 copy/validation——三个任务互不改对方文件，靠这份清单衔接。
+    # 缺失模块/类在装配时报 CONFIG_ERROR 并列出缺失项，绝不静默跳过。
+    report_pipeline_stages: list[str] = Field(
+        default_factory=lambda: [
+            "packages.report.vision_stage:VisionStage",
+            "packages.report.diagnosis_stage:DiagnosisStage",
+            "packages.report.copy_stage:CopyStage",
+            "packages.report.validation_stage:ValidationStage",
+        ],
+        alias="REPORT_PIPELINE_STAGES_JSON",
+    )
+    # 视觉提供者：deterministic = V1 骨架的确定性假实现（真实 RTMPose/BiRefNet
+    # 是 Sprint 1 结论落地后的替换项，接口位见 packages/report/vision.py）。
+    report_vision_provider: str = "deterministic"
+    report_tier: str = "basic"
+
     # --- Observability -----------------------------------------------------
     otel_enabled: bool = False
     otel_service_name: str = "meida-api"
@@ -228,6 +246,7 @@ class Settings(BaseSettings):
         "task_bonus_rules",
         "upload_scenes",
         "retrieval_rerank_weights",
+        "report_pipeline_stages",
         mode="before",
     )
     @classmethod