Update base for Update on "Refactor use_triton_kernel to use nvfp4_quantize_kernel_choice"

Summary:
This is to prefer the addition of flashinfer quantize kernel path in next PR

Test Plan:
python test/prototype/mx_formats/test_inference_workflow.py

Reviewers:

Subscribers:

Tasks:

Tags:

[ghstack-poisoned]

Author: jerryzh168

test/quantization/pt2e/test_x86inductor_fusion.py

@@ -3093,13 +3093,19 @@ def matcher_check_fn():
 
     @skipIfNoDynamoSupport
     @skipIfNoONEDNN
+    @unittest.skipIf(
+        torch_version_at_least("2.11.0.dev"), "Doesn't work with torch 2.11.0.dev+"
+    )
     def test_q_attention_block(self):
         for annotate_matmul in [True, False]:
             self._test_q_attention_block_helper(annotate_matmul=annotate_matmul)
 
     @skipIfNoDynamoSupport
     @skipIfNoONEDNN
     @skipIfNoFloat8Support
+    @unittest.skipIf(
+        torch_version_at_least("2.11.0.dev"), "Doesn't work with torch 2.11.0.dev+"
+    )
     def test_fp8_q_attention_block(self):
         for annotate_matmul in [True, False]:
             self._test_q_attention_block_helper(