Apply ruff formatting fixes

Signed-off-by: Mert Toslali <[email protected]>

Author: toslali-ibm

src/guidellm/backends/response_handlers.py

@@ -146,7 +146,7 @@ def compile_non_streaming(
             request_args=str(
                 request.arguments.model_dump() if request.arguments else None
             ),
-            response_id=response.get("id"), # use vLLM ID if available
+            response_id=response.get("id"),  # use vLLM ID if available
             text=text,
             input_metrics=input_metrics,
             output_metrics=output_metrics,
@@ -196,7 +196,7 @@ def compile_streaming(self, request: GenerationRequest) -> GenerationResponse:
             request_args=str(
                 request.arguments.model_dump() if request.arguments else None
             ),
-            response_id=self.streaming_response_id, # use vLLM ID if available
+            response_id=self.streaming_response_id,  # use vLLM ID if available
             text=text,
             input_metrics=input_metrics,
             output_metrics=output_metrics,
@@ -312,7 +312,7 @@ def compile_non_streaming(
             request_args=str(
                 request.arguments.model_dump() if request.arguments else None
             ),
-            response_id=response.get("id"), # use vLLM ID if available
+            response_id=response.get("id"),  # use vLLM ID if available
             text=text,
             input_metrics=input_metrics,
             output_metrics=output_metrics,
@@ -362,7 +362,7 @@ def compile_streaming(self, request: GenerationRequest) -> GenerationResponse:
             request_args=str(
                 request.arguments.model_dump() if request.arguments else None
             ),
-            response_id=self.streaming_response_id, # use vLLM ID if available
+            response_id=self.streaming_response_id,  # use vLLM ID if available
             text=text,
             input_metrics=input_metrics,
             output_metrics=output_metrics,
@@ -398,7 +398,6 @@ def __init__(self):
         self.streaming_usage: dict[str, int | dict[str, int]] | None = None
         self.streaming_response_id: str | None = None
 
-
     def compile_non_streaming(
         self, request: GenerationRequest, response: dict
     ) -> GenerationResponse:
@@ -421,7 +420,7 @@ def compile_non_streaming(
             request_args=str(
                 request.arguments.model_dump() if request.arguments else None
             ),
-            response_id=response.get("id"), # use vLLM ID if available
+            response_id=response.get("id"),  # use vLLM ID if available
             text=text,
             input_metrics=input_metrics,
             output_metrics=output_metrics,

src/guidellm/schemas/response.py

@@ -43,7 +43,7 @@ class GenerationResponse(StandardBaseModel):
     )
     response_id: str | None = Field(
         default=None,
-        description="Unique identifier matching the original vLLM Response ID."
+        description="Unique identifier matching the original vLLM Response ID.",
     )
     request_args: str | None = Field(
         description="Arguments passed to the backend for request processing."