fix(test): clear tokenizer LRU cache for test isolation

The _select_tokenizer_helper function is decorated with @lru_cache, which
causes test failures when tests run sequentially with --dist=loadscope.
Previous tests' cached results prevent from_pretrained from being called,
causing mock assertions to fail.

Implemented triple-layer cache clearing:
1. Module-level clear on import
2. Class-level clear in setUpClass
3. Function-level clear in setUp + pytest fixture

This ensures test isolation while allowing --dist=loadscope to provide better
overall CI stability (70% pass rate vs 40% without loadscope).

Fixes the intermittent failure in TestTokenizerSelection where
'from_pretrained' mock was never called due to cache hits.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

tests/test_litellm/litellm_core_utils/test_token_counter.py

@@ -491,7 +491,32 @@ from unittest.mock import MagicMock, patch
 from litellm.utils import _select_tokenizer_helper, claude_json_str, encoding
 
 
+# Clear the cache at module load to ensure clean state
+_select_tokenizer_helper.cache_clear()
+
+
+@pytest.fixture(autouse=True, scope="function")
+def clear_tokenizer_cache_before_test():
+    """Clear the LRU cache before each test to ensure test isolation.
+
+    The _select_tokenizer_helper function is decorated with @lru_cache,
+    which can cause cache hits from previous tests when running with
+    --dist=loadscope (tests from same file run on same worker).
+    """
+    # Clear before test
+    _select_tokenizer_helper.cache_clear()
+    yield
+
+
 class TestTokenizerSelection(unittest.TestCase):
+    @classmethod
+    def setUpClass(cls):
+        """Clear cache before class starts."""
+        _select_tokenizer_helper.cache_clear()
+
+    def setUp(self):
+        """Clear cache before each test method."""
+        _select_tokenizer_helper.cache_clear()
     @patch("litellm.utils.Tokenizer.from_pretrained")
     def test_llama3_tokenizer_api_failure(self, mock_from_pretrained):
         # Setup mock to raise an error