refactor: always call validate_environment() instead of using flag-based approach

- Remove requires_custom_headers_validation() method from BaseConfig
- Remove requires_custom_headers_validation() override from GithubCopilotConfig
- Always call validate_environment() in OpenAI completion flow
- Remove related tests for the removed flag method

Author: krozbicki

litellm/llms/base_llm/chat/transformation.py

@@ -154,18 +154,6 @@ def should_fake_stream(
         """
         return False
 
-    def requires_custom_headers_validation(self) -> bool:
-        """
-        Returns True if the provider needs validate_environment to be called
-        to add custom headers before making the request.
-        
-        This is used for OpenAI-compatible providers that require special headers
-        (e.g., GitHub Copilot needs Copilot-Vision-Request for vision requests).
-        
-        Override this method in provider-specific config classes that need it.
-        """
-        return False
-
     def _add_tools_to_optional_params(self, optional_params: dict, tools: List) -> dict:
         """
         Helper util to add tools to optional_params.

litellm/llms/github_copilot/chat/transformation.py

@@ -20,14 +20,6 @@ def __init__(
         super().__init__()
         self.authenticator = Authenticator()
 
-    def requires_custom_headers_validation(self) -> bool:
-        """
-        GitHub Copilot requires custom headers validation to add:
-        - X-Initiator header based on message roles
-        - Copilot-Vision-Request header for vision requests
-        """
-        return True
-
     def _get_openai_compatible_provider_info(
         self,
         model: str,

litellm/llms/openai/openai.py

@@ -554,9 +554,9 @@ def completion(  # type: ignore # noqa: PLR0915
                     model=model, custom_llm_provider=custom_llm_provider, stream=stream
                 )
 
-            # Validate environment and update headers for providers that require it
+            # Always call validate_environment to allow providers to add custom headers
             # (e.g., GitHub Copilot needs Copilot-Vision-Request for vision requests)
-            if provider_config and provider_config.requires_custom_headers_validation():
+            if provider_config:
                 validated_headers = provider_config.validate_environment(
                     headers=headers or {},
                     model=model or "",

tests/test_litellm/llms/github_copilot/test_github_copilot_transformation.py

@@ -509,22 +509,3 @@ def test_copilot_vision_request_header_with_type_image_url():
 
     assert headers["Copilot-Vision-Request"] == "true"
     assert headers["X-Initiator"] == "user"
-
-
-def test_requires_custom_headers_validation():
-    """Test that GithubCopilotConfig.requires_custom_headers_validation() returns True"""
-    config = GithubCopilotConfig()
-    
-    # GitHub Copilot should require custom headers validation
-    assert config.requires_custom_headers_validation() is True
-
-
-def test_base_config_requires_custom_headers_validation():
-    """Test that BaseConfig.requires_custom_headers_validation() returns False by default"""
-    from litellm.llms.openai.chat.gpt_transformation import OpenAIGPTConfig
-    
-    # Use a concrete implementation that doesn't override requires_custom_headers_validation
-    config = OpenAIGPTConfig()
-    
-    # Default implementation should return False
-    assert config.requires_custom_headers_validation() is False