feat(api): 叙事 prompt、职业上下文、读路径章节、WS 解耦与错误脱敏

- 回忆录：事实边界补充允许清单；传记文体示例与 JSON 叙事要求对齐
- default 职业提示 occupation_context；cadre/military 退休语境
- GET 章节读路径零写入，prepare_chapter_read_view + markdown_for_response
- 文本归一抽到 core/text_normalize；移除弃用 reply 策略与 recompose_chapters_for_story
- ConversationService：WS 连接/用户段落/结束对话；对外错误固定文案
- 测试：HTTP 脱敏契约、章节读视图、occupation 与 background_voice

api/app/agents/chat/interview_reply_length.py

@@ -309,49 +309,3 @@ def bump_reply_plan_for_background_voice(
         likely_chit_chat=plan.likely_chit_chat,
         information_rich=plan.information_rich,
     )
-
-
-# 向后兼容：旧名与旧签名（仅测试或外部引用）
-def compute_reply_length_strategy(
-    user_message_len: int,
-    *,
-    likely_new_detail: bool,
-    likely_chit_chat: bool,
-    settings: "Settings",
-) -> ReplyPlan:
-    """已弃用：请用 compute_reply_plan(user_message, ...)。保留供过渡期。"""
-    # 无法还原 information_rich，按旧逻辑近似
-    n = max(0, int(user_message_len))
-    max_segments = int(settings.chat_interview_max_segments)
-    if n <= _LEN_BRIEF_MAX:
-        mode = ReplyLengthMode.brief
-    elif n < _LEN_LONG_MIN:
-        mode = ReplyLengthMode.standard
-    else:
-        if likely_chit_chat:
-            mode = ReplyLengthMode.standard
-        elif likely_new_detail:
-            mode = ReplyLengthMode.expanded
-        else:
-            mode = ReplyLengthMode.standard
-    return _plan_from_mode(
-        mode,
-        max_segments=max_segments,
-        settings=settings,
-        background_voice=None,
-        likely_new=likely_new_detail,
-        likely_chit=likely_chit_chat,
-        info_rich=False,
-    )
-
-
-def bump_reply_length_strategy_for_background_voice(
-    plan: ReplyPlan,
-    *,
-    background_voice: str | None,
-    settings: "Settings",
-) -> ReplyPlan:
-    """旧名兼容。"""
-    return bump_reply_plan_for_background_voice(
-        plan, background_voice=background_voice, settings=settings
-    )