refactor(chat): AI-native prompts, remove interview heuristics

- Drop interview_reply_length and utterance_substance; always run stage LLM
  and memory retrieval when enabled; trim Settings fields and .env.example.
- Replace guided/opening prompts with compact fact blocks plus unified
  behavior guidance; slim background_voice and persona to tone hints.
- InterviewAgent uses fixed chat_interview max_tokens/chars/segments.

Also includes stacked work: profile followup/extract path, evaluation rubric
and judge schema updates, transcript SPLIT handling in execution service,
user export markdown split tests, and golden case fixture.

api/app/agents/chat/output_rules.py

@@ -8,7 +8,9 @@ def chat_output_rules() -> str:
         "反引号代码、`[]()` 链接、列表符号或渲染用符号；只输出连贯口语，**可以**在需要分两气泡时使用字面量 "
         "`[SPLIT]`（仅此一处方括号用法）；**禁止**输出括号、括号内的策略/舞台说明（例如「（先接住情绪）」「（共情）」）、"
         "思考过程或任何元注释——这些只存在于系统指令里，**绝不可**出现在你对用户说的话中；"
-        "采访腔（「我注意到」「我想了解」）；重复确认对方已经说过或能推断出的信息；编造对方没说的细节。"
+        "采访腔（「我注意到」「我想了解」）；重复确认对方已经说过或能推断出的信息；"
+        "编造对方没说的**具体**事实（人名、时间、地点、事件经过等若用户未提及则不说）。"
+        "**允许**用「我能想象……」「那时候大概……」等泛泛接话，但不要把这些写成就等于用户亲身经历的事实。"
     )