Convert benchmarks to ruff format (#18068)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

benchmarks/kernels/graph_machete_bench.py

@@ -14,13 +14,14 @@ from vllm.utils import FlexibleArgumentParser
 
 if __name__ == "__main__":
     parser = FlexibleArgumentParser(
-        description='Benchmark the latency of processing a single batch of '
-        'requests till completion.')
-    parser.add_argument('filename', type=str)
+        description="Benchmark the latency of processing a single batch of "
+        "requests till completion."
+    )
+    parser.add_argument("filename", type=str)
 
     args = parser.parse_args()
 
-    with open(args.filename, 'rb') as f:
+    with open(args.filename, "rb") as f:
         data = pickle.load(f)
         raw_results: list[TMeasurement] = data["results"]
 
@@ -38,11 +39,7 @@ if __name__ == "__main__":
             raise Exception("MKN not found")
 
         kernel = v.task_spec.description
-        results[KN].append({
-            "kernel": kernel,
-            "batch_size": M,
-            "median": v.median
-        })
+        results[KN].append({"kernel": kernel, "batch_size": M, "median": v.median})
 
     rows = int(math.ceil(len(results) / 2))
     fig, axs = plt.subplots(rows, 2, figsize=(12, 5 * rows))
@@ -50,14 +47,16 @@ if __name__ == "__main__":
     for axs_idx, (shape, data) in enumerate(results.items()):
         plt.sca(axs[axs_idx])
         df = pd.DataFrame(data)
-        sns.lineplot(data=df,
-                     x="batch_size",
-                     y="median",
-                     hue="kernel",
-                     style="kernel",
-                     markers=True,
-                     dashes=False,
-                     palette="Dark2")
+        sns.lineplot(
+            data=df,
+            x="batch_size",
+            y="median",
+            hue="kernel",
+            style="kernel",
+            markers=True,
+            dashes=False,
+            palette="Dark2",
+        )
         plt.title(f"Shape: {shape}")
         plt.ylabel("time (median, s)")
     plt.tight_layout()