feat: split API and provider specs into separate llama-stack-api pkg

Extract API definitions, models, and provider specifications into a
standalone llama-stack-api package that can be published to PyPI
independently of the main llama-stack server.

Motivation

External providers currently import from llama-stack, which overrides
the installed version and causes dependency conflicts. This separation
allows external providers to:

- Install only the type definitions they need without server dependencies
- Avoid version conflicts with the installed llama-stack package
- Be versioned and released independently

This enables us to re-enable external provider module tests that were
previously blocked by these import conflicts.

Changes

- Created llama-stack-api package with minimal dependencies (pydantic, jsonschema)
- Moved APIs, providers datatypes, strong_typing, and schema_utils
- Updated all imports from llama_stack.* to llama_stack_api.*
- Preserved git history using git mv for moved files
- Configured local editable install for development workflow
- Updated linting and type-checking configuration for both packages
- Rebased on top of upstream src/ layout changes

Testing

Package builds successfully and can be imported independently.
All pre-commit hooks pass with expected exclusions maintained.

Next Steps

- Publish llama-stack-api to PyPI
- Update external provider dependencies
- Re-enable external provider module tests

Signed-off-by: Charlie Doern <cdoern@redhat.com>

tests/unit/providers/vector_io/conftest.py

@@ -9,9 +9,9 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import numpy as np
 import pytest
+from llama_stack_api.apis.vector_io import Chunk, ChunkMetadata, QueryChunksResponse
+from llama_stack_api.apis.vector_stores import VectorStore
 
-from llama_stack.apis.vector_io import Chunk, ChunkMetadata, QueryChunksResponse
-from llama_stack.apis.vector_stores import VectorStore
 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

tests/unit/providers/vector_io/test_faiss.py

@@ -9,11 +9,11 @@ from unittest.mock import MagicMock, patch
 
 import numpy as np
 import pytest
+from llama_stack_api.apis.files import Files
+from llama_stack_api.apis.vector_io import Chunk, QueryChunksResponse
+from llama_stack_api.apis.vector_stores import VectorStore
+from llama_stack_api.providers.datatypes import HealthStatus
 
-from llama_stack.apis.files import Files
-from llama_stack.apis.vector_io import Chunk, QueryChunksResponse
-from llama_stack.apis.vector_stores import VectorStore
-from llama_stack.providers.datatypes import HealthStatus
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.inline.vector_io.faiss.faiss import (
     FaissIndex,

tests/unit/providers/vector_io/test_sqlite_vec.py

@@ -8,8 +8,8 @@ import asyncio
 
 import numpy as np
 import pytest
+from llama_stack_api.apis.vector_io import Chunk, QueryChunksResponse
 
-from llama_stack.apis.vector_io import Chunk, QueryChunksResponse
 from llama_stack.providers.inline.vector_io.sqlite_vec.sqlite_vec import (
     SQLiteVecIndex,
     SQLiteVecVectorIOAdapter,

tests/unit/providers/vector_io/test_vector_io_openai_vector_stores.py

@@ -10,9 +10,8 @@ from unittest.mock import AsyncMock, patch
 
 import numpy as np
 import pytest
-
-from llama_stack.apis.common.errors import VectorStoreNotFoundError
-from llama_stack.apis.vector_io import (
+from llama_stack_api.apis.common.errors import VectorStoreNotFoundError
+from llama_stack_api.apis.vector_io import (
     Chunk,
     OpenAICreateVectorStoreFileBatchRequestWithExtraBody,
     OpenAICreateVectorStoreRequestWithExtraBody,
@@ -20,7 +19,8 @@ from llama_stack.apis.vector_io import (
     VectorStoreChunkingStrategyAuto,
     VectorStoreFileObject,
 )
-from llama_stack.apis.vector_stores import VectorStore
+from llama_stack_api.apis.vector_stores import VectorStore
+
 from llama_stack.providers.inline.vector_io.sqlite_vec.sqlite_vec import VECTOR_DBS_PREFIX
 
 # This test is a unit test for the inline VectorIO providers. This should only contain
@@ -222,7 +222,7 @@ async def test_insert_chunks_missing_db_raises(vector_io_adapter):
 
 async def test_insert_chunks_with_missing_document_id(vector_io_adapter):
     """Ensure no KeyError when document_id is missing or in different places."""
-    from llama_stack.apis.vector_io import Chunk, ChunkMetadata
+    from llama_stack_api.apis.vector_io import Chunk, ChunkMetadata
 
     fake_index = AsyncMock()
     vector_io_adapter.cache["db1"] = fake_index
@@ -255,7 +255,7 @@ async def test_insert_chunks_with_missing_document_id(vector_io_adapter):
 
 async def test_document_id_with_invalid_type_raises_error():
     """Ensure TypeError is raised when document_id is not a string."""
-    from llama_stack.apis.vector_io import Chunk
+    from llama_stack_api.apis.vector_io import Chunk
 
     # Integer document_id should raise TypeError
     from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id

tests/unit/providers/vector_io/test_vector_utils.py

@@ -4,7 +4,8 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.vector_io import Chunk, ChunkMetadata
+from llama_stack_api.apis.vector_io import Chunk, ChunkMetadata
+
 from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id
 
 # This test is a unit test for the chunk_utils.py helpers. This should only contain