[Refactor] Simplify `chat_completion_full_generator` for tool parsers (#35634)

Signed-off-by: yewentao256 <zhyanwentao@126.com>

vllm/entrypoints/openai/chat_completion/serving.py

@@ -1463,17 +1463,7 @@ class OpenAIServingChat(OpenAIServing):
             tool_call_class = (
                 MistralToolCall if is_mistral_tokenizer(tokenizer) else ToolCall
             )
-            if self.use_harmony:
-                # Harmony models already have parsed content and tool_calls
-                # through parse_chat_output. Respect its output directly.
-                message = ChatMessage(
-                    role=role,
-                    reasoning=reasoning,
-                    content=content,
-                    tool_calls=tool_calls if tool_calls else [],
-                )
-
-            elif (not self.enable_auto_tools or not self.tool_parser) and (
+            if (not self.enable_auto_tools or not self.tool_parser) and (
                 not isinstance(request.tool_choice, ChatCompletionNamedToolChoiceParam)
                 and request.tool_choice != "required"
             ):