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/features/conversation/ws/router.py

@@ -24,11 +24,11 @@ from app.features.conversation.ws.message_types import MessageType
 from app.features.conversation.ws.pipeline import (
     _delayed_listening_feedback,
     _voice_session_id_from_client_segment_id,
-    background_runner,
     bump_tts_cancel_epoch,
     chat_orchestrator,
     cleanup_segment_states,
     get_or_create_segment_state,
+    memoir_ingest_scheduler,
     process_audio_segment,
     process_conversation_segments,
     process_user_message,
@@ -304,7 +304,7 @@ async def websocket_endpoint(
                                 text_message,
                             )
                             user_message_timestamp = conversation.last_message_at
-                            await background_runner.queue_message(
+                            await memoir_ingest_scheduler.queue_segment(
                                 conversation.user_id,
                                 segment.id,
                                 text_char_count=len(text_message.strip()),
@@ -565,7 +565,7 @@ async def websocket_endpoint(
                                     )
                                 )
                                 user_message_timestamp = conversation.last_message_at
-                                await background_runner.queue_message(
+                                await memoir_ingest_scheduler.queue_segment(
                                     conversation.user_id,
                                     segment.id,
                                     text_char_count=len((asr_text or "").strip()),