fix: remove importlib.reload calls that cause cross-test class-reference staleness

Two test files were reloading modules in setup_method/fixtures, which
caused class-reference staleness for subsequent tests in the same worker:

1. test_huggingface_embedding_handler.py reloaded
   litellm.llms.custom_httpx.http_handler, creating a new HTTPHandler
   class. Subsequent tests (e.g. hosted_vllm embedding) created
   client = HTTPHandler() from the new class, but llm_http_handler.py
   still held the old class reference. isinstance(client, HTTPHandler)
   returned False, so a new unpatched client was used and
   client.post was never called.

2. test_vertex_ai_rerank_integration.py reloaded
   litellm.llms.vertex_ai.rerank.transformation in setup_method,
   creating a new VertexAIRerankConfig class. The transformation test
   file's module-level import still referenced the old class, so
   @patch('...VertexAIRerankConfig._ensure_access_token') patched the
   new class while self.config was an instance of the old class,
   leaving the mock unapplied and hitting real Google credentials.

Fix: remove the reload calls. The module-level class references are
stable across tests within a worker; the reloads were solving a problem
that doesn't exist and actively created cross-test contamination.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/test_litellm/llms/huggingface/embedding/test_huggingface_embedding_handler.py

@@ -1,4 +1,3 @@
-import importlib
 import json
 import os
 import sys
@@ -16,22 +15,7 @@ MOCK_EMBEDDING_RESPONSE = [[0.1, 0.2, 0.3, 0.4, 0.5]]
 
 
 @pytest.fixture
-def reload_huggingface_modules():
-    """
-    Reload modules to ensure fresh references after conftest reloads litellm.
-    This ensures the HTTPHandler class being patched is the same one used by
-    the embedding handler during parallel test execution.
-    """
-    import litellm.llms.custom_httpx.http_handler as http_handler_module
-    import litellm.llms.huggingface.embedding.handler as hf_embedding_handler_module
-
-    importlib.reload(http_handler_module)
-    importlib.reload(hf_embedding_handler_module)
-    yield
-
-
-@pytest.fixture
-def mock_embedding_http_handler(reload_huggingface_modules):
+def mock_embedding_http_handler():
     """Fixture to mock the HTTP handler for embedding tests"""
     with patch("litellm.llms.custom_httpx.http_handler.HTTPHandler.post") as mock_post:
         mock_response = MagicMock()
@@ -43,7 +27,7 @@ def mock_embedding_http_handler(reload_huggingface_modules):
 
 
 @pytest.fixture
-def mock_embedding_async_http_handler(reload_huggingface_modules):
+def mock_embedding_async_http_handler():
     """Fixture to mock the async HTTP handler for embedding tests"""
     with patch("litellm.llms.custom_httpx.http_handler.AsyncHTTPHandler.post", new_callable=AsyncMock) as mock_post:
         mock_response = MagicMock()

tests/test_litellm/llms/vertex_ai/rerank/test_vertex_ai_rerank_integration.py

@@ -2,7 +2,6 @@
 Integration tests for Vertex AI rerank functionality.
 These tests demonstrate end-to-end usage of the Vertex AI rerank feature.
 """
-import importlib
 import os
 from unittest.mock import MagicMock, patch
 
@@ -14,14 +13,7 @@ from litellm.llms.vertex_ai.rerank.transformation import VertexAIRerankConfig
 
 class TestVertexAIRerankIntegration:
     def setup_method(self):
-        # Reload modules to ensure fresh references after conftest reloads litellm.
-        # This ensures the class being patched is the same one used by the tests.
-        import litellm.llms.vertex_ai.rerank.transformation as rerank_transformation_module
-        importlib.reload(rerank_transformation_module)
-
-        # Re-import after reload to get the fresh class
-        from litellm.llms.vertex_ai.rerank.transformation import VertexAIRerankConfig as FreshConfig
-        self.config = FreshConfig()
+        self.config = VertexAIRerankConfig()
         self.model = "semantic-ranker-default@latest"
 
     @patch('litellm.llms.vertex_ai.rerank.transformation.VertexAIRerankConfig._ensure_access_token')