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/agents/memoir/story_route_payload.py

@@ -7,6 +7,7 @@ Story 路由：候选故事 JSON 载荷（summary 优先、预算裁剪、固定
 from __future__ import annotations
 
 import json
+import re
 from datetime import timezone
 from typing import Any, TYPE_CHECKING
 
@@ -15,6 +16,23 @@ if TYPE_CHECKING:
 
 from app.features.story.models import Story
 
+_PLAIN_SNIPPET_NOISE = re.compile(r"[`*_#]+")
+
+
+def _plain_opening_snippet_from_markdown(md: str, *, max_chars: int) -> str:
+    """无 summary 时供路由辨题的短文摘（弱化 Markdown 噪声）。"""
+    t = (md or "").strip()
+    if not t:
+        return ""
+    t = re.sub(r"!\[[^\]]*\]\([^)]+\)", "", t)
+    t = re.sub(r"\[([^\]]+)\]\([^)]+\)", r"\1", t)
+    t = re.sub(r"asset://\S+", "", t)
+    t = _PLAIN_SNIPPET_NOISE.sub("", t)
+    t = re.sub(r"\s+", " ", t).strip()
+    if len(t) <= max_chars:
+        return t
+    return t[: max_chars - 1] + "…"
+
 
 def _linked_chapters(s: Story) -> list[str]:
     links: list[str] = []
@@ -126,6 +144,9 @@ def _build_full_row(
     )
     if body:
         row["body_for_route"] = body
+    osnip = _plain_opening_snippet_from_markdown(canon, max_chars=260)
+    if osnip and len(osnip) >= 40:
+        row["opening_snippet"] = osnip
     return row