fix: ensure production uses DB-configured model/provider from UI

- Migration 026: update seed defaults from ollama to vllm/AxionML
- Migration 031: fix existing rows still on old ollama defaults
- Helm values: set OLLAMA_BASE_URL to cluster ollama endpoint (was empty)
- Extractor: guard against switching to ollama when base_url is empty
- OllamaClient: validate base_url on construction to fail fast

services/extractor/client.py

@@ -140,6 +140,11 @@ class OllamaClient:
         max_retries: int | None = None,
         http_client: httpx.AsyncClient | None = None,
     ) -> None:
+        if not config.base_url or not config.base_url.startswith(("http://", "https://")):
+            raise ValueError(
+                f"OllamaClient requires a valid base_url (got {config.base_url!r}). "
+                "Set OLLAMA_BASE_URL environment variable."
+            )
         self._config = config
         self._max_retries = max_retries if max_retries is not None else config.max_retries
         self._base_delay = config.retry_base_delay

services/extractor/main.py

@@ -486,17 +486,25 @@ async def main() -> None:
                         model_changed = new_cfg.model != extractor_client._config.model
 
                         if provider_changed or model_changed:
-                            logger.info(
-                                "Extractor provider switch: old_provider=%s new_provider=%s "
-                                "model=%s variant=%s",
-                                old_provider, new_provider,
-                                new_resolved.model_name, new_resolved.variant_id,
-                            )
-                            await extractor_client.close()
-                            extractor_client = build_llm_client(
-                                new_resolved, config.ollama, config.vllm,
-                            )
-                            extractor_provider = new_provider
+                            # Guard: don't switch to ollama if base_url is empty
+                            if new_provider == "ollama" and not config.ollama.base_url:
+                                logger.warning(
+                                    "DB resolved provider=ollama but OLLAMA_BASE_URL is empty — "
+                                    "keeping current %s client. Fix the agent config in the UI.",
+                                    extractor_provider,
+                                )
+                            else:
+                                logger.info(
+                                    "Extractor provider switch: old_provider=%s new_provider=%s "
+                                    "model=%s variant=%s",
+                                    old_provider, new_provider,
+                                    new_resolved.model_name, new_resolved.variant_id,
+                                )
+                                await extractor_client.close()
+                                extractor_client = build_llm_client(
+                                    new_resolved, config.ollama, config.vllm,
+                                )
+                                extractor_provider = new_provider
                         else:
                             # Same provider and model — just update config in-place
                             extractor_client._config = new_cfg  # type: ignore[assignment]
@@ -517,18 +525,26 @@ async def main() -> None:
                         cls_model_changed = new_cls_cfg.model != classifier_client._config.model
 
                         if cls_provider_changed or cls_model_changed:
-                            logger.info(
-                                "Classifier provider switch: old_provider=%s new_provider=%s "
-                                "model=%s variant=%s",
-                                old_cls_provider, new_cls_provider,
-                                new_cls_resolved.model_name, new_cls_resolved.variant_id,
-                            )
-                            if classifier_client is not extractor_client:
-                                await classifier_client.close()
-                            classifier_client = build_llm_client(
-                                new_cls_resolved, config.ollama, config.vllm,
-                            )
-                            classifier_provider = new_cls_provider
+                            # Guard: don't switch to ollama if base_url is empty
+                            if new_cls_provider == "ollama" and not config.ollama.base_url:
+                                logger.warning(
+                                    "DB resolved classifier provider=ollama but OLLAMA_BASE_URL is empty — "
+                                    "keeping current %s client. Fix the agent config in the UI.",
+                                    classifier_provider,
+                                )
+                            else:
+                                logger.info(
+                                    "Classifier provider switch: old_provider=%s new_provider=%s "
+                                    "model=%s variant=%s",
+                                    old_cls_provider, new_cls_provider,
+                                    new_cls_resolved.model_name, new_cls_resolved.variant_id,
+                                )
+                                if classifier_client is not extractor_client:
+                                    await classifier_client.close()
+                                classifier_client = build_llm_client(
+                                    new_cls_resolved, config.ollama, config.vllm,
+                                )
+                                classifier_provider = new_cls_provider
                         elif classifier_client is extractor_client and new_cls_cfg.model != extractor_client._config.model:
                             classifier_client = build_llm_client(
                                 new_cls_resolved, config.ollama, config.vllm,