[MXFP4] Add 1D TMA copy for Scale tensor in MXFP4 GEMM (#766)

* [TMA] Add 1D TMA copy for Scale tensor

* [Lint]

* [Test] Add test for kernel

* [BugFix]

examples/dequantize_gemm/test_example_dequantize_gemm.py

@@ -3,6 +3,7 @@ import tilelang.testing
 import example_dequant_gemv_fp16xint4
 import example_dequant_gemm_fp4_hopper
 import example_dequant_gemm_bf16_mxfp4_hopper
+import example_dequant_gemm_bf16_mxfp4_hopper_tma
 
 
 @tilelang.testing.requires_cuda
@@ -22,5 +23,11 @@ def test_example_dequant_gemm_bf16_mxfp4_hopper():
     example_dequant_gemm_bf16_mxfp4_hopper.main()
 
 
+@tilelang.testing.requires_cuda
+@tilelang.testing.requires_cuda_compute_version_ge(9, 0)
+def test_example_dequant_gemm_bf16_mxfp4_hopper_tma():
+    example_dequant_gemm_bf16_mxfp4_hopper_tma.main()
+
+
 if __name__ == "__main__":
     tilelang.testing.main()