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?

app-expo/tests/features/conversation/message-split.test.ts

@@ -1,5 +1,7 @@
 import {
+  assistantSegmentMessageId,
   lastSegmentPreview,
+  normalizeAssistantContentForSplit,
   splitMessageParts,
   splitStreamingSegments,
 } from '@/features/conversation/message-split';
@@ -11,11 +13,31 @@ describe('message-split', () => {
     expect(splitMessageParts('a [Split] b')).toEqual(['a', 'b']);
   });
 
+  it('splitMessageParts handles spaces inside brackets and fullwidth brackets', () => {
+    expect(splitMessageParts('a [ SPLIT ] b')).toEqual(['a', 'b']);
+    expect(splitMessageParts('a［SPLIT］b')).toEqual(['a', 'b']);
+    expect(splitMessageParts('a【SPLIT】b')).toEqual(['a', 'b']);
+  });
+
   it('splitMessageParts trims and drops empty segments', () => {
     expect(splitMessageParts('  x  [SPLIT]  y  ')).toEqual(['x', 'y']);
     expect(splitMessageParts('[SPLIT]only')).toEqual(['only']);
   });
 
+  it('splitMessageParts splits multi-segment content as persisted / WS-joined', () => {
+    expect(splitMessageParts('第一段[SPLIT]第二段')).toEqual([
+      '第一段',
+      '第二段',
+    ]);
+  });
+
+  it('splitMessageParts falls back to double-newline paragraphs (no [SPLIT] in DB)', () => {
+    const a = '太为你高兴了！在上海大剧院的舞台绽放，聚光灯下的你。';
+    const b =
+      '说到舞台，我忽然想起你黄浦江边的童年。从看着江水流淌，到在舞台上演绎别人的悲欢。';
+    expect(splitMessageParts(`${a}\n\n${b}`)).toEqual([a, b]);
+  });
+
   it('splitStreamingSegments keeps empty tail after delimiter', () => {
     expect(splitStreamingSegments('first [SPLIT]')).toEqual(['first', '']);
   });
@@ -24,4 +46,14 @@ describe('message-split', () => {
     expect(lastSegmentPreview('a [SPLIT] b', 10)).toBe('b');
     expect(lastSegmentPreview('hello', 3)).toBe('hel');
   });
+
+  it('assistantSegmentMessageId matches WS / TTS segment binding', () => {
+    expect(assistantSegmentMessageId('uuid-a', 0)).toBe('uuid-a_seg_0');
+    expect(assistantSegmentMessageId('uuid-a', 1)).toBe('uuid-a_seg_1');
+  });
+
+  it('normalizeAssistantContentForSplit maps fullwidth brackets', () => {
+    expect(normalizeAssistantContentForSplit('［x］')).toBe('[x]');
+    expect(normalizeAssistantContentForSplit('【x】')).toBe('[x]');
+  });
 });