feat(api)!: memory single chain — async MemoryService, strict eval closure

Route all memory ingest/retrieve/enrichment/compaction through async MemoryService.
Remove legacy sync memory implementations (ingest/retrieve/compaction); Celery and
memoir Phase2 call asyncio.run into MemoryService-backed helpers.

Memoir Phase1 batch ingest uses MemoryService.ingest_transcripts_batch; drop chapters.
evidence_bundle_json mirror (Alembic 0015). Evaluation uses snapshot/link-only bundles;
raise EvidenceClosureMissing instead of partial/fallback lineage tiers.

Split memoir state into NarrativeCoverageState and InterviewControlState; delete the
_interview_meta_store adapter layer. Remove rolling-query and recent-fact fallback
settings from config and evidence assembly.

Update judges, docs, tests, and PlaygroundPage alignment.

Made-with: Cursor

api/app/agents/chat/interview_agent.py

@@ -31,7 +31,11 @@ from app.agents.chat.reply_limits import (
 )
 from app.agents.chat.reply_planner import maybe_refine_turn_plan_with_llm
 from app.agents.chat.stage_detection import keyword_fallback_primary_stage
-from app.agents.state_schema import MemoirStateSchema
+from app.agents.state_schema import (
+    MemoirStateSchema,
+    interview_control_state,
+    narrative_coverage_state,
+)
 from app.core.agent_logging import (
     agent_span,
     log_agent_payload,
@@ -154,14 +158,14 @@ class InterviewAgent:
             text_for_model = self._resolve_text_for_model(
                 user_message, normalized_user_message
             )
-            empty_slots = memoir_state.prompt_empty_slots_for_current_stage()
-            filled_slots = {
-                key: value.snippet
-                for key, value in memoir_state.slots.get(
-                    memoir_state.current_stage, {}
-                ).items()
-                if value.snippet
-            }
+            narrative_state = narrative_coverage_state(memoir_state)
+            control_state = interview_control_state(memoir_state)
+            empty_slots = control_state.prompt_empty_slots_for_stage(
+                narrative_state, memoir_state.current_stage
+            )
+            filled_slots = narrative_state.filled_slots_for_stage(
+                memoir_state.current_stage
+            )
             if detected_user_stage is not None:
                 du = detected_user_stage
             else:
@@ -173,7 +177,7 @@ class InterviewAgent:
             )
             recent_questions = extract_recent_questions(hw.window)
             conversation_turn_total = hw.turn_total
-            all_stages_coverage = memoir_state.all_stages_coverage()
+            all_stages_coverage = narrative_state.all_stages_coverage()
             persona = normalize_interview_persona(settings.chat_interview_persona)
             max_segments = int(settings.chat_interview_max_segments)
             max_tokens = int(settings.chat_interview_max_tokens)
@@ -406,7 +410,11 @@ class InterviewAgent:
         if not self.llm:
             return ["你好呀～ 又见面了。今天想从人生里哪一小段回忆开始聊聊？"]
         try:
-            empty_slots = memoir_state.prompt_empty_slots_for_current_stage()
+            narrative_state = narrative_coverage_state(memoir_state)
+            control_state = interview_control_state(memoir_state)
+            empty_slots = control_state.prompt_empty_slots_for_stage(
+                narrative_state, memoir_state.current_stage
+            )
             empty_slots_readable = [SLOT_NAME_MAP.get(s, s) for s in empty_slots]
             persona = normalize_interview_persona(settings.chat_interview_persona)
             prompt = get_opening_prompt(