fix: Use re-entrancy and concurrency safe context managers for provider data (#1498)

Concurrent requests should not trample (or reuse) each others' provider
data. Provider data should be scoped to each request.

## Test Plan

Set the uvicorn server to have a single worker process + thread by
updating the config:
```python
    uvicorn_config = {
        ...
        "workers": 1,
        "loop": "asyncio",
    }
```

Then perform the following steps on `origin/main` (without this change).

(1) Run the server using `llama stack run dev` without having
`FIREWORKS_API_KEY` in the environment.

(2) Run a test by specifying the FIREWORKS_API_KEY env var so it gets
stored in the thread local
```
pytest -s -v tests/integration/inference/test_text_inference.py \
    --stack-config http://localhost:8321 \
    --text-model accounts/fireworks/models/llama-v3p1-8b-instruct \
    -k test_text_chat_completion_with_tool_calling_and_streaming \
     --env FIREWORKS_API_KEY=<...>
``` 
Ensure you don't have any other API keys in the environment (otherwise
the bug will not reproduce due to other specifics in our testing code.)
Verify this works.

(3) Run the same command again without specifying FIREWORKS_API_KEY. See
that the request actually succeeds when it *should have failed*.


----
Now do the same tests on this branch, verify step (3) results in
failure.

Finally, run the full `test_text_inference.py` test suite with this
change, verify it succeeds.

llama_stack/distribution/library_client.py

@@ -32,7 +32,10 @@ from termcolor import cprint
 from llama_stack.distribution.build import print_pip_install_help
 from llama_stack.distribution.configure import parse_and_maybe_upgrade_config
 from llama_stack.distribution.datatypes import Api
-from llama_stack.distribution.request_headers import set_request_provider_data
+from llama_stack.distribution.request_headers import (
+    preserve_headers_context_async_generator,
+    request_provider_data_context,
+)
 from llama_stack.distribution.resolver import ProviderRegistry
 from llama_stack.distribution.server.endpoints import get_all_api_endpoints
 from llama_stack.distribution.stack import (
@@ -262,21 +265,25 @@ class AsyncLlamaStackAsLibraryClient(AsyncLlamaStackClient):
         if not self.endpoint_impls:
             raise ValueError("Client not initialized")
 
+        # Create headers with provider data if available
+        headers = {}
         if self.provider_data:
-            set_request_provider_data({"X-LlamaStack-Provider-Data": json.dumps(self.provider_data)})
+            headers["X-LlamaStack-Provider-Data"] = json.dumps(self.provider_data)
 
-        if stream:
-            response = await self._call_streaming(
-                cast_to=cast_to,
-                options=options,
-                stream_cls=stream_cls,
-            )
-        else:
-            response = await self._call_non_streaming(
-                cast_to=cast_to,
-                options=options,
-            )
-        return response
+        # Use context manager for provider data
+        with request_provider_data_context(headers):
+            if stream:
+                response = await self._call_streaming(
+                    cast_to=cast_to,
+                    options=options,
+                    stream_cls=stream_cls,
+                )
+            else:
+                response = await self._call_non_streaming(
+                    cast_to=cast_to,
+                    options=options,
+                )
+            return response
 
     def _find_matching_endpoint(self, method: str, path: str) -> tuple[Any, dict]:
         """Find the matching endpoint implementation for a given method and path.
@@ -374,9 +381,12 @@ class AsyncLlamaStackAsLibraryClient(AsyncLlamaStackClient):
             finally:
                 await end_trace()
 
+        # Wrap the generator to preserve context across iterations
+        wrapped_gen = preserve_headers_context_async_generator(gen())
+
         mock_response = httpx.Response(
             status_code=httpx.codes.OK,
-            content=gen(),
+            content=wrapped_gen,
             headers={
                 "Content-Type": "application/json",
             },