feat(api): 收敛对话与记忆流程边界，引入 LLM 网关与专用服务

- MemoryService 异步路径委托 MemoryIngestService / MemoryRetrievalService；富化派发经 MemoryEnrichmentScheduler
- WebSocket pipeline 经 ChatTurnService 与显式 DTO 编排单轮对话；回忆录片段入队由 MemoirIngestScheduler 封装
- 新增 LlmGateway（LlmUseCase），各 agent、任务与适配器对齐 ports
- 补充 memory 提示适配、runtime 类型、memory-retrieval 文档、ai-touchpoints 说明与扫描脚本及配套测试

Made-with: Cursor

api/app/tasks/memoir_quality_pass_tasks.py

@@ -18,7 +18,7 @@ from sqlalchemy.orm import Session
 from app.agents.memoir.narrative_agent import NarrativeAgent
 from app.core.config import settings
 from app.core.db import get_sync_db
-from app.core.dependencies import get_llm_provider
+from app.core.llm_gateway import LlmGateway, LlmUseCase
 from app.core.logging import get_logger
 from app.core.memoir_pipeline_progress import merge_pipeline_run
 from app.features.memoir.models import Chapter
@@ -30,7 +30,7 @@ logger = get_logger(__name__)
 
 def _get_llm():
     try:
-        return getattr(get_llm_provider(), "langchain_llm", None)
+        return LlmGateway().langchain_llm_for(LlmUseCase("memoir_quality_pass"))
     except Exception:
         return None
 

api/app/tasks/memoir_tasks.py

@@ -27,7 +27,7 @@ from app.core.chapter_pipeline_lock import (
 )
 from app.core.config import settings
 from app.core.db import get_sync_db
-from app.core.dependencies import get_llm_provider, get_llm_provider_fast
+from app.core.llm_gateway import LlmGateway, LlmUseCase
 from app.core.logging import get_logger
 from app.core.memoir_pipeline_progress import (
     init_pipeline_run_from_phase1,
@@ -129,7 +129,7 @@ def _run_post_pipeline_commit(
 def _get_llm():
     """Celery 任务内获取 LangChain LLM（通过 port）"""
     try:
-        return getattr(get_llm_provider(), "langchain_llm", None)
+        return LlmGateway().langchain_llm_for(LlmUseCase("memoir_tasks"))
     except Exception:
         return None
 
@@ -137,7 +137,9 @@ def _get_llm():
 def _get_llm_fast():
     """分类 / 抽取等快档位任务（与叙事、路由默认模型可分离）。"""
     try:
-        return getattr(get_llm_provider_fast(), "langchain_llm", None)
+        return LlmGateway().langchain_llm_for(
+            LlmUseCase("memoir_tasks.fast", fast=True)
+        )
     except Exception:
         return None
 

api/app/tasks/story_title_tasks.py

@@ -5,7 +5,7 @@ import time
 from celery import shared_task
 
 from app.core.db import get_sync_db
-from app.core.dependencies import get_llm_provider
+from app.core.llm_gateway import LlmGateway, LlmUseCase
 from app.core.logging import get_logger
 
 logger = get_logger(__name__)
@@ -62,7 +62,7 @@ def generate_story_title_after_create(
                 )
                 return {"status": "skip_user_modified"}
 
-            llm = getattr(get_llm_provider(), "langchain_llm", None)
+            llm = LlmGateway().langchain_llm_for(LlmUseCase("story_title"))
             if not llm:
                 ms = (time.perf_counter() - t0) * 1000
                 logger.info(