fix: remove sentence-transformers from remote vllm

vLLM itself can perform the embeddings generation so we don't need this
extra provider.

Signed-off-by: Sébastien Han <seb@redhat.com>

llama_stack/templates/remote-vllm/run-with-safety.yaml

@@ -26,9 +26,6 @@ providers:
       max_tokens: ${env.VLLM_MAX_TOKENS:4096}
       api_token: ${env.VLLM_API_TOKEN:fake}
       tls_verify: ${env.VLLM_TLS_VERIFY:true}
-  - provider_id: sentence-transformers
-    provider_type: inline::sentence-transformers
-    config: {}
   vector_io:
   - provider_id: faiss
     provider_type: inline::faiss
@@ -133,7 +130,7 @@ models:
 - metadata:
     embedding_dimension: 384
   model_id: all-MiniLM-L6-v2
-  provider_id: sentence-transformers
+  provider_id: vllm-inference
   model_type: embedding
 shields:
 - shield_id: ${env.SAFETY_MODEL}