Fix Dell distro and TGI inference

llama_stack/distributions/dell/run.yaml

@@ -23,6 +23,9 @@ providers:
     provider_type: remote::chromadb
     config:
       url: ${env.CHROMA_URL}
+      kvstore:
+        type: sqlite
+        db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/starter/}/chroma_remote_registry.db
   safety:
   - provider_id: llama-guard
     provider_type: inline::llama-guard

llama_stack/providers/remote/inference/tgi/tgi.py

@@ -309,7 +309,7 @@ class TGIAdapter(_HfAdapter):
             raise ValueError("You must provide a URL in run.yaml (or via the TGI_URL environment variable) to use TGI.")
         log.info(f"Initializing TGI client with url={config.url}")
         self.client = AsyncInferenceClient(
-            model=config.url,
+            model=config.url,provider="hf-inference"
         )
         endpoint_info = await self.client.get_endpoint_info()
         self.max_tokens = endpoint_info["max_total_tokens"]