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/tests/test_user_export_markdown_split.py

@@ -0,0 +1,24 @@
+"""导出 Markdown 中 AI 块的 [SPLIT] 规范化。"""
+
+from app.features.evaluation.importers.user_export_markdown import (
+    extract_dialogue_turns_from_export_md,
+)
+
+
+def test_extract_dialogue_turns_replaces_split_in_ai():
+    md = """
+#### 轮次 1
+
+**用户:**
+你好。
+
+**AI:**
+第一段[SPLIT]第二段
+"""
+    turns = extract_dialogue_turns_from_export_md(md)
+    assert len(turns) == 1
+    u, a = turns[0]
+    assert u == "你好。"
+    assert "[SPLIT]" not in a
+    assert "第一段" in a and "第二段" in a
+    assert "\n" in a