fix: honor hyde_enabled toggle and record ping failures before client creation (Task 6 review)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

backend/app/api/routes/status.py

@@ -169,6 +169,13 @@ def _build_model_status(role: str) -> dict[str, Any]:
             # Config always wins: report "disabled" even if the reranker was
             # enabled and called successfully earlier in this process's life.
             status = "disabled"
+    elif role == "hyde_llm":
+        enabled = settings.hyde_enabled
+        if not enabled:
+            # Same "config always wins" override as the reranker branch above:
+            # report "disabled" even if HyDE ran successfully before being
+            # turned off in settings during this process's life.
+            status = "disabled"
 
     return {
         "role": role,
@@ -199,7 +206,15 @@ async def get_model_statuses():
 async def _ping_main_or_hyde(role: str) -> None:
     """Send one minimal chat completion to the LLM configured for `role`."""
     provider, model = _resolve_role_provider_model(role)
-    client = get_llm_client(provider=provider, model=model)
+    try:
+        client = get_llm_client(provider=provider, model=model)
+    except Exception as exc:  # noqa: BLE001 - record, then re-raise so gather() still isolates this ping
+        # get_llm_client() can fail before any TrackedLLMClient exists to
+        # record the outcome itself (e.g. missing API key, unsupported
+        # provider string), so record the failure here directly, otherwise it
+        # would be invisible on the /status/models page afterward.
+        get_model_usage_tracker().record(provider=provider, model=model, success=False, error=str(exc))
+        raise
     await asyncio.to_thread(client.chat, [{"role": "user", "content": "ping"}], max_tokens=1)