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/features/memoir/story_pipeline_sync.py

@@ -36,7 +36,6 @@ from app.agents.stage_constants import (
 )
 from app.agents.state_schema import MemoirStateSchema
 from app.core.config import settings
-from app.core.dependencies import get_embedding_provider
 from app.core.logging import get_logger
 from app.features.conversation.lineage_schemas import aggregate_lineage_from_segments
 from app.features.memoir.chapter_evidence_snapshot import (
@@ -60,7 +59,6 @@ from app.features.memoir.repo import (
     mark_chapter_dirty_sync,
     reorder_chapter_story_links_by_life_order_sync,
 )
-from app.features.memory.repo import retrieve_evidence_sync
 from app.features.story.models import Story, StoryVersion
 from app.features.story.sync_write import (
     append_story_version_sync,
@@ -102,7 +100,7 @@ def _dialogue_lineage_dict_for_segment_ids(
 def _evidence_link_ids(
     evidence: dict,
 ) -> tuple[list[str], list[str], list[str], list[str]]:
-    """从 retrieve_evidence_sync 结果提取稳定 ID 列表。"""
+    """从 MemoryService.retrieve 结果提取稳定 ID 列表。"""
     chunks: list[str] = []
     for c in evidence.get("relevant_chunks") or []:
         if isinstance(c, dict) and c.get("id"):
@@ -960,6 +958,7 @@ def run_story_pipeline_for_category_batch(
     occupation: str = "",
     memoir_correlation_id: str | None = None,
     llm_fast: Any | None = None,
+    memory_evidence: dict | None = None,
 ) -> tuple[Chapter | None, bool, set[str]]:
     """
     返回 (chapter, needs_cover_enqueue, story_ids_to_dispatch_after_commit)。
@@ -979,8 +978,6 @@ def run_story_pipeline_for_category_batch(
     top_k = int(settings.evidence_top_k_default)
     if n_units > int(settings.evidence_large_batch_threshold):
         top_k = int(settings.evidence_top_k_large_batch)
-    emb = get_embedding_provider()
-    embedding_available = emb.is_available()
 
     def _oral_job() -> tuple[str, float]:
         t_oral = time.perf_counter()
@@ -991,23 +988,13 @@ def run_story_pipeline_for_category_batch(
     with ThreadPoolExecutor(max_workers=1) as pool:
         oral_future = pool.submit(_oral_job)
         _t_ev = time.perf_counter()
-        try:
-            evidence = retrieve_evidence_sync(
-                session,
-                user_id,
-                combined_text,
-                top_k=top_k,
-                embedding_provider=emb,
-            )
-        except Exception as e:
-            logger.warning("Evidence 检索跳过: {}", e)
-            evidence = {
-                "relevant_chunks": [],
-                "relevant_summaries": [],
-                "relevant_facts": [],
-                "timeline_hints": [],
-                "relevant_stories": [],
-            }
+        evidence = memory_evidence or {
+            "relevant_chunks": [],
+            "relevant_summaries": [],
+            "relevant_facts": [],
+            "timeline_hints": [],
+            "relevant_stories": [],
+        }
         ev_elapsed = time.perf_counter() - _t_ev
         oral_for_memoir, oral_elapsed = oral_future.result()
     pipeline_phase_timings["evidence"] = ev_elapsed
@@ -1017,13 +1004,13 @@ def run_story_pipeline_for_category_batch(
     )
 
     logger.info(
-        "memoir_evidence_retrieved user_id={} chunks={} facts={} summaries={} stories={} vector_ok={}",
+        "memoir_evidence_retrieved user_id={} chunks={} facts={} summaries={} stories={} top_k={}",
         user_id,
         len(evidence.get("relevant_chunks") or []),
         len(evidence.get("relevant_facts") or []),
         len(evidence.get("relevant_summaries") or []),
         len(evidence.get("relevant_stories") or []),
-        embedding_available,
+        top_k,
     )
 
     evidence_text = format_evidence_chunks_for_prompt(evidence)