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/alembic/versions/0015_memory_single_chain_cleanup.py

@@ -0,0 +1,86 @@
+"""Memory single-chain cleanup and memoir state control columns.
+
+Revision ID: 0015_memory_single_chain
+Revises: 0014_memory_evidence_indexes
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+from alembic import op
+
+revision: str = "0015_memory_single_chain"
+down_revision: Union[str, None] = "0014_memory_evidence_indexes"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def _has_column(table: str, column: str) -> bool:
+    bind = op.get_bind()
+    return any(c["name"] == column for c in sa.inspect(bind).get_columns(table))
+
+
+def upgrade() -> None:
+    if not _has_column("memoir_states", "known_facts_json"):
+        op.add_column(
+            "memoir_states",
+            sa.Column(
+                "known_facts_json",
+                postgresql.JSON(astext_type=sa.Text()),
+                nullable=True,
+            ),
+        )
+    if not _has_column("memoir_states", "persona_threads_json"):
+        op.add_column(
+            "memoir_states",
+            sa.Column(
+                "persona_threads_json",
+                postgresql.JSON(astext_type=sa.Text()),
+                nullable=True,
+            ),
+        )
+    if not _has_column("memoir_states", "recent_questions_json"):
+        op.add_column(
+            "memoir_states",
+            sa.Column(
+                "recent_questions_json",
+                postgresql.JSON(astext_type=sa.Text()),
+                nullable=True,
+            ),
+        )
+
+    op.execute(
+        """
+        UPDATE memoir_states
+        SET
+          known_facts_json = COALESCE((slots::jsonb -> '__interview_state__' -> 'known_facts')::json, '[]'::json),
+          persona_threads_json = COALESCE((slots::jsonb -> '__interview_state__' -> 'persona_threads')::json, '[]'::json),
+          recent_questions_json = COALESCE((slots::jsonb -> '__interview_state__' -> 'recent_questions')::json, '[]'::json),
+          slots = (slots::jsonb - '__interview_state__')::json
+        WHERE slots::jsonb ? '__interview_state__'
+        """
+    )
+
+    if _has_column("chapters", "evidence_bundle_json"):
+        op.drop_column("chapters", "evidence_bundle_json")
+
+
+def downgrade() -> None:
+    if not _has_column("chapters", "evidence_bundle_json"):
+        op.add_column(
+            "chapters",
+            sa.Column(
+                "evidence_bundle_json",
+                postgresql.JSON(astext_type=sa.Text()),
+                nullable=True,
+            ),
+        )
+    for column in (
+        "recent_questions_json",
+        "persona_threads_json",
+        "known_facts_json",
+    ):
+        if _has_column("memoir_states", column):
+            op.drop_column("memoir_states", column)