feat(evaluation): memoir readiness, judge/replay updates, eval web playground

Add memoir_readiness_service and router tests; extend judge schemas/services, replay_service, and conversation rubric; align story route agent, payload, prompts, and story_pipeline_sync; update agent logging, config, and DI. Document internal-eval; add replayDraft util and PlaygroundPage changes in app-eval-web.

api/app/features/evaluation/router.py

@@ -14,6 +14,7 @@ from app.features.evaluation.admin_service import EvaluationAdminService
 from app.features.evaluation.deps import (
     get_eval_judge_manual_service,
     get_evaluation_admin_service,
+    get_memoir_readiness_service,
     get_replay_conversation_service,
 )
 from app.features.evaluation.errors import (
@@ -27,6 +28,7 @@ from app.features.evaluation.importers.user_export_markdown import (
 from app.features.evaluation.internal_auth import InternalEvalAuth
 from app.features.evaluation.judge_manual_service import EvalJudgeManualService
 from app.features.evaluation.presenters import case_out, run_out
+from app.features.evaluation.memoir_readiness_service import MemoirReadinessService
 from app.features.evaluation.replay_service import ReplayConversationService
 from app.features.evaluation.schemas import (
     CaseCreate,
@@ -49,6 +51,7 @@ from app.features.evaluation.schemas import (
     RegressionSetOut,
     ReplayBootstrapBody,
     ReplayBootstrapOut,
+    MemoirPhase1ReadyOut,
     ReplayConversationBody,
     ReplayConversationOut,
     SessionDialogueOut,
@@ -222,6 +225,35 @@ async def get_session_transcript(
     )
 
 
+@router.get(
+    "/sessions/{conversation_id}/memoir-phase1-ready",
+    response_model=MemoirPhase1ReadyOut,
+)
+async def memoir_phase1_ready(
+    conversation_id: str,
+    _auth: InternalEvalAuth,
+    svc: Annotated[
+        MemoirReadinessService, Depends(get_memoir_readiness_service)
+    ],
+    segment_ids: Annotated[
+        list[str],
+        Query(
+            min_length=1,
+            description="本批待检查的 segment id，可重复 query 参数 segment_ids=id1&segment_ids=id2",
+        ),
+    ],
+):
+    try:
+        return await svc.memoir_phase1_ready_for_segments(
+            conversation_id=conversation_id,
+            segment_ids=segment_ids,
+        )
+    except EvaluationNotFoundError as e:
+        raise _eval_http_exc(e) from e
+    except EvaluationBadRequestError as e:
+        raise _eval_http_exc(e) from e
+
+
 @router.get(
     "/sessions/{conversation_id}/evaluation-runs",
     response_model=SessionEvalRunsOut,
@@ -282,9 +314,10 @@ async def replay_conversation(
             detail="provide only one of fixture_filename or user_utterances",
         )
     try:
+        segment_ids: list[str] = []
         if body.fixture_filename:
             fn = body.fixture_filename.strip()
-            n, echo = await replay.replay_fixture(
+            n, echo, segment_ids = await replay.replay_fixture(
                 conversation_id=body.conversation_id,
                 fixture_filename=fn,
                 flush_memoir_after=body.flush_memoir_after,
@@ -294,7 +327,7 @@ async def replay_conversation(
             utt = [str(u) for u in body.user_utterances if str(u).strip()]
             if not utt:
                 raise EvaluationBadRequestError("user_utterances is empty")
-            n = await replay.replay_utterances(
+            n, segment_ids = await replay.replay_utterances(
                 conversation_id=body.conversation_id,
                 utterances=utt,
                 flush_memoir_after=body.flush_memoir_after,
@@ -313,6 +346,7 @@ async def replay_conversation(
         conversation_id=body.conversation_id,
         turns_replayed=n,
         utterances_echo=echo,
+        segment_ids=segment_ids,
     )