feat(stores)!: use backend storage references instead of configs (#3697)

**This PR changes configurations in a backward incompatible way.**

Run configs today repeat full SQLite/Postgres snippets everywhere a
store is needed, which means duplicated credentials, extra connection
pools, and lots of drift between files. This PR introduces named storage
backends so the stack and providers can share a single catalog and
reference those backends by name.

## Key Changes

- Add `storage.backends` to `StackRunConfig`, register each KV/SQL
backend once at startup, and validate that references point to the right
family.
- Move server stores under `storage.stores` with lightweight references
(backend + namespace/table) instead of full configs.
- Update every provider/config/doc to use the new reference style;
docs/codegen now surface the simplified YAML.

## Migration

Before:
```yaml
metadata_store:
  type: sqlite
  db_path: ~/.llama/distributions/foo/registry.db
inference_store:
  type: postgres
  host: ${env.POSTGRES_HOST}
  port: ${env.POSTGRES_PORT}
  db: ${env.POSTGRES_DB}
  user: ${env.POSTGRES_USER}
  password: ${env.POSTGRES_PASSWORD}
conversations_store:
  type: postgres
  host: ${env.POSTGRES_HOST}
  port: ${env.POSTGRES_PORT}
  db: ${env.POSTGRES_DB}
  user: ${env.POSTGRES_USER}
  password: ${env.POSTGRES_PASSWORD}
```

After:
```yaml
storage:
  backends:
    kv_default:
      type: kv_sqlite
      db_path: ~/.llama/distributions/foo/kvstore.db
    sql_default:
      type: sql_postgres
      host: ${env.POSTGRES_HOST}
      port: ${env.POSTGRES_PORT}
      db: ${env.POSTGRES_DB}
      user: ${env.POSTGRES_USER}
      password: ${env.POSTGRES_PASSWORD}
  stores:
    metadata:
      backend: kv_default
      namespace: registry
    inference:
      backend: sql_default
      table_name: inference_store
      max_write_queue_size: 10000
      num_writers: 4
    conversations:
      backend: sql_default
      table_name: openai_conversations
```

Provider configs follow the same pattern—for example, a Chroma vector
adapter switches from:

```yaml
providers:
  vector_io:
  - provider_id: chromadb
    provider_type: remote::chromadb
    config:
      url: ${env.CHROMADB_URL}
      kvstore:
        type: sqlite
        db_path: ~/.llama/distributions/foo/chroma.db
```

to:

```yaml
providers:
  vector_io:
  - provider_id: chromadb
    provider_type: remote::chromadb
    config:
      url: ${env.CHROMADB_URL}
      persistence:
        backend: kv_default
        namespace: vector_io::chroma_remote
```

Once the backends are declared, everything else just points at them, so
rotating credentials or swapping to Postgres happens in one place and
the stack reuses a single connection pool.

tests/unit/providers/vector_io/conftest.py

@@ -12,13 +12,14 @@ import pytest
 
 from llama_stack.apis.vector_dbs import VectorDB
 from llama_stack.apis.vector_io import Chunk, ChunkMetadata, QueryChunksResponse
+from llama_stack.core.storage.datatypes import KVStoreReference, SqliteKVStoreConfig
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.inline.vector_io.faiss.faiss import FaissIndex, FaissVectorIOAdapter
 from llama_stack.providers.inline.vector_io.sqlite_vec import SQLiteVectorIOConfig
 from llama_stack.providers.inline.vector_io.sqlite_vec.sqlite_vec import SQLiteVecIndex, SQLiteVecVectorIOAdapter
 from llama_stack.providers.remote.vector_io.pgvector.config import PGVectorVectorIOConfig
 from llama_stack.providers.remote.vector_io.pgvector.pgvector import PGVectorIndex, PGVectorVectorIOAdapter
-from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
+from llama_stack.providers.utils.kvstore import register_kvstore_backends
 
 EMBEDDING_DIMENSION = 768
 COLLECTION_PREFIX = "test_collection"
@@ -112,8 +113,9 @@ async def unique_kvstore_config(tmp_path_factory):
     unique_id = f"test_kv_{np.random.randint(1e6)}"
     temp_dir = tmp_path_factory.getbasetemp()
     db_path = str(temp_dir / f"{unique_id}.db")
-
-    return SqliteKVStoreConfig(db_path=db_path)
+    backend_name = f"kv_vector_{unique_id}"
+    register_kvstore_backends({backend_name: SqliteKVStoreConfig(db_path=db_path)})
+    return KVStoreReference(backend=backend_name, namespace=f"vector_io::{unique_id}")
 
 
 @pytest.fixture(scope="session")
@@ -138,7 +140,7 @@ async def sqlite_vec_vec_index(embedding_dimension, tmp_path_factory):
 async def sqlite_vec_adapter(sqlite_vec_db_path, unique_kvstore_config, mock_inference_api, embedding_dimension):
     config = SQLiteVectorIOConfig(
         db_path=sqlite_vec_db_path,
-        kvstore=unique_kvstore_config,
+        persistence=unique_kvstore_config,
     )
     adapter = SQLiteVecVectorIOAdapter(
         config=config,
@@ -177,7 +179,7 @@ async def faiss_vec_index(embedding_dimension):
 @pytest.fixture
 async def faiss_vec_adapter(unique_kvstore_config, mock_inference_api, embedding_dimension):
     config = FaissVectorIOConfig(
-        kvstore=unique_kvstore_config,
+        persistence=unique_kvstore_config,
     )
     adapter = FaissVectorIOAdapter(
         config=config,
@@ -253,7 +255,7 @@ async def pgvector_vec_adapter(unique_kvstore_config, mock_inference_api, embedd
         db="test_db",
         user="test_user",
         password="test_password",
-        kvstore=unique_kvstore_config,
+        persistence=unique_kvstore_config,
     )
 
     adapter = PGVectorVectorIOAdapter(config, mock_inference_api, None)