feat(bot): chat agent on Gemini 2.5 Flash-Lite — grounded first turn, current-prescription defaults

Live-tested on OpenRouter: Flash-Lite answers in 3–8 s with correct findings at ~$0.10/$0.40 per 1M tokens
(qwen3.7-flash and gpt-oss-120b took 26–32 s). Force a tool call on the first turn, give the model the
current prescription, let check_drug_for_patient default to it and return composition, intolerance check
and general interactions (alcohol); retry once on an empty final message.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

app/llm/agent/providers.py

@@ -52,7 +52,10 @@ class OpenRouterProvider:
         self._client = client
 
     async def chat(self, messages: list[dict[str, Any]], tools: list[dict[str, Any]]) -> AssistantTurn:
-        body = {"model": self._model, "messages": messages, "tools": tools, "temperature": 0.1, "max_tokens": 900}
+        body: dict[str, Any] = {"model": self._model, "messages": messages, "tools": tools, "temperature": 0.1}
+        body["max_tokens"] = 900
+        if not any(m["role"] == "tool" for m in messages):
+            body["tool_choice"] = "required"  # never answer from the model's own memory: ground first
         headers = {"Authorization": f"Bearer {self._key}", "X-Title": "PTSRaiAgent prototype (synthetic data)"}
         try:
             if self._client is not None: