feat(memory,conversation): 记忆富化/证据包、时间线幂等字段与对话分段全链路

数据库
- 新增迁移 0003：timeline_events.memory_source_id 外键 → memory_sources，便于按 ingest 源做时间线幂等

后端 - 记忆
- 新增 ingest 后 LLM 富化（摘要/事实/时间线），可配置开关与最大字符数
- 新增证据包组装：合并 chunk、摘要、事实、时间线、故事等检索结果；支持空 query 时是否仍带 rolling 等开关
- repo/retriever/service/router/schemas/summarizer/timeline/extractor 等扩展；文档 memory-retrieval.md 更新

后端 - 对话 WS
- 增加 PING/PONG；分段 ASR 日志与空音频处理；转写失败与「无助手回复」错误提示更明确
- 助手多段回复持久化使用统一分隔符，与分段逻辑一致

后端 - Agent
- reply_limits：按 [SPLIT] 与段落拆段，并保证非空 fallback，供 WS 与 TTS 多段下发

后端 - 回忆录任务
- transcript ingest 记录 source_id；任务成功结?

api/app/agents/chat/interview_agent.py

@@ -17,7 +17,11 @@ from app.agents.chat.prompts_conversation import (
     get_opening_prompt,
 )
 from app.agents.state_schema import MemoirStateSchema
-from app.agents.chat.reply_limits import truncate_chat_segments
+from app.agents.chat.reply_limits import (
+    nonempty_segments_or_fallback,
+    segments_from_llm_response,
+    truncate_chat_segments,
+)
 from app.core.agent_logging import (
     agent_span,
     log_agent_payload,
@@ -135,10 +139,12 @@ class InterviewAgent:
             log_agent_payload(
                 logger, "InterviewAgent.generate_response.raw_response", response_text
             )
-            messages = [
-                msg.strip() for msg in response_text.split("[SPLIT]") if msg.strip()
-            ]
-            raw_list = messages if messages else [response_text.strip()]
+            raw_list = segments_from_llm_response(
+                response_text,
+                max_segments=settings.chat_interview_max_segments,
+            )
+            if not raw_list:
+                raw_list = [response_text.strip()]
             out = truncate_chat_segments(
                 raw_list,
                 max_segments=settings.chat_interview_max_segments,
@@ -150,6 +156,7 @@ class InterviewAgent:
                         : settings.chat_interview_max_chars_per_segment
                     ]
                 ]
+            out = nonempty_segments_or_fallback(out, fallback=_FALLBACK_REPLY)
             log_agent_summary(
                 logger,
                 "InterviewAgent.generate_response segments={} conversation_id={}",
@@ -193,10 +200,9 @@ class InterviewAgent:
             log_agent_payload(
                 logger, "InterviewAgent.opening.raw_response", response_text
             )
-            messages = [
-                msg.strip() for msg in response_text.split("[SPLIT]") if msg.strip()
-            ]
-            raw_list = messages if messages else [response_text.strip()]
+            raw_list = segments_from_llm_response(response_text, max_segments=2)
+            if not raw_list:
+                raw_list = [response_text.strip()]
             out = truncate_chat_segments(
                 raw_list,
                 max_segments=2,
@@ -208,7 +214,7 @@ class InterviewAgent:
                 len(out),
                 conversation_id,
             )
-            return (
+            segments = (
                 out
                 if out
                 else [
@@ -217,6 +223,10 @@ class InterviewAgent:
                     ]
                 ]
             )
+            return nonempty_segments_or_fallback(
+                segments,
+                fallback="你好呀～ 又见面了，最近有没有什么事想跟我说说？",
+            )
         except Exception as e:
             logger.error("生成开场白失败: {}", e, exc_info=True)
             return ["你好呀～ 又见面了，最近有没有什么事想跟我说说？"]