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/profile_agent.py

@@ -152,6 +152,12 @@ class ProfileAgent:
                 result["grew_up_place"] = str(parsed.grew_up_place)
             if parsed.occupation:
                 result["occupation"] = str(parsed.occupation)
+            bp = result.get("birth_place")
+            gp = result.get("grew_up_place")
+            if bp and not gp:
+                result["grew_up_place"] = bp
+            elif gp and not bp:
+                result["birth_place"] = gp
             return result
         except Exception as e:
             logger.error("提取资料信息失败: {}", e)
@@ -173,7 +179,6 @@ class ProfileAgent:
             prompt = get_profile_followup_prompt(
                 missing_fields,
                 filled_fields,
-                user_message,
                 nickname,
                 interview_stage_hint=interview_stage_hint,
             )