feat(chat): persona decides to buy/walk — auto-finish + tolerance (temper) + resume

The conversation now ENDS when the persona makes a decision (option C), not when the
trainee clicks a button:
- persona_id replies carry {reply, decision(none/buy/walk), mood}; when decision is
  buy/walk the session auto-finishes (won/lost) with a debrief that reveals latent
  details + per-turn 'turning points'.
- personas have a tolerance (1-5, 'temper'): impatient personas walk away fast after
  poor answers (fed via internal.misses on mood<=-1); tough 'wrong text' cases can
  still be won by a strong, gentle response (judge realism).
- trainee 'Finish' button removed; if they leave mid-chat an active session is resumed
  via /chat/resume (continue, not restart). One-shot lock still enforced once decided.
- mock/tests updated: persona deciding buy -> send auto-finishes won.
Rebuilt dist.

backend/scripts/mock_llm.py

@@ -107,5 +107,10 @@ class MockLLM:
         return {}
 
     def complete_conversation(self, messages, **kw) -> str:
-        # persona chat: echo a short in-character reply
-        return json.dumps({"reply": "I see. Tell me more about the price then."}, ensure_ascii=False)
+        # persona chat: echo a short in-character reply with a decision.
+        # On the first send, the persona decides to buy (so E2E auto-finishes as won).
+        return json.dumps({
+            "reply": "I see. Tell me more about the price then.",
+            "decision": "buy",
+            "mood": 1,
+        }, ensure_ascii=False)