[sgl-kernel][Feat][B200][1/N] Support MXFP8 Grouped GEMM in Blackwell (#13731)

Co-authored-by: Yineng Zhang <me@zhyncs.com>

sgl-kernel/include/sgl_kernel_ops.h

@@ -909,6 +909,24 @@ void es_fp8_blockwise_scaled_grouped_mm(
     const torch::Tensor& expert_offsets,
     const torch::Tensor& workspace);
 
+void es_sm100_mxfp8_blockscaled_grouped_mm(
+    const torch::Tensor& a,
+    const torch::Tensor& b,
+    const torch::Tensor& sfa,
+    const torch::Tensor& sfb,
+    torch::Tensor& d,
+    const torch::Tensor& problem_sizes,
+    const torch::Tensor& expert_offsets,
+    const torch::Tensor& blockscale_offsets);
+
+void es_sm100_mxfp8_blockscaled_grouped_quant(
+    const torch::Tensor& input,
+    const torch::Tensor& problem_sizes,
+    const torch::Tensor& expert_offsets,
+    const torch::Tensor& blockscale_offsets,
+    torch::Tensor& quant_output,
+    torch::Tensor& scale_factor);
+
 /*
  * From fast-hadamard-transform
  */