fix: actualize chunking strategy in vector store create API

# What does this PR do?


## Test Plan

src/llama_stack/core/routers/vector_io.py

@@ -20,6 +20,8 @@ from llama_stack.apis.vector_io import (
     SearchRankingOptions,
     VectorIO,
     VectorStoreChunkingStrategy,
+    VectorStoreChunkingStrategyStatic,
+    VectorStoreChunkingStrategyStaticConfig,
     VectorStoreDeleteResponse,
     VectorStoreFileBatchObject,
     VectorStoreFileContentsResponse,
@@ -167,6 +169,13 @@ class VectorIORouter(VectorIO):
         if embedding_dimension is not None:
             params.model_extra["embedding_dimension"] = embedding_dimension
 
+        # Set chunking strategy explicitly if not provided
+        if params.chunking_strategy is None or params.chunking_strategy.type == "auto":
+            # actualize the chunking strategy to static
+            params.chunking_strategy = VectorStoreChunkingStrategyStatic(
+                static=VectorStoreChunkingStrategyStaticConfig()
+            )
+
         return await provider.openai_create_vector_store(params)
 
     async def openai_list_vector_stores(

src/llama_stack/providers/utils/memory/openai_vector_store_mixin.py

@@ -26,6 +26,7 @@ from llama_stack.apis.vector_io import (
     VectorStoreChunkingStrategy,
     VectorStoreChunkingStrategyAuto,
     VectorStoreChunkingStrategyStatic,
+    VectorStoreChunkingStrategyStaticConfig,
     VectorStoreContent,
     VectorStoreDeleteResponse,
     VectorStoreFileBatchObject,
@@ -414,6 +415,10 @@ class OpenAIVectorStoreMixin(ABC):
             in_progress=0,
             total=0,
         )
+        if not params.chunking_strategy or params.chunking_strategy.type == "auto":
+            chunking_strategy = VectorStoreChunkingStrategyStatic(static=VectorStoreChunkingStrategyStaticConfig())
+        else:
+            chunking_strategy = params.chunking_strategy
         store_info: dict[str, Any] = {
             "id": vector_store_id,
             "object": "vector_store",
@@ -426,7 +431,7 @@ class OpenAIVectorStoreMixin(ABC):
             "expires_at": None,
             "last_active_at": created_at,
             "file_ids": [],
-            "chunking_strategy": params.chunking_strategy,
+            "chunking_strategy": chunking_strategy.model_dump(),
         }
 
         # Add provider information to metadata if provided

tests/integration/vector_io/test_openai_vector_stores.py

@@ -679,7 +679,7 @@ def test_openai_vector_store_attach_file(
     assert file_attach_response.id == file.id
     assert file_attach_response.vector_store_id == vector_store.id
     assert file_attach_response.status == "completed"
-    assert file_attach_response.chunking_strategy.type == "auto"
+    assert file_attach_response.chunking_strategy.type == "static"
     assert file_attach_response.created_at > 0
     assert not file_attach_response.last_error
 
@@ -815,8 +815,8 @@ def test_openai_vector_store_list_files(
     assert set(file_ids) == {file.id for file in files_list.data}
     assert files_list.data[0].object == "vector_store.file"
     assert files_list.data[0].vector_store_id == vector_store.id
-    assert files_list.data[0].status == "completed"
-    assert files_list.data[0].chunking_strategy.type == "auto"
+    assert files_list.data[0].status in ["completed", "in_progress"]
+    assert files_list.data[0].chunking_strategy.type == "static"
     assert files_list.data[0].created_at > 0
     assert files_list.first_id == files_list.data[0].id
     assert not files_list.data[0].last_error