Fix compilation errors generated by forthcoming Clang changes (#1544)

Without this change, the following diagnostic is generated:
  a template argument list is expected after a name prefixed by the template
  keyword [-Wmissing-template-arg-list-after-template-kw]

See C++17 spec [temp.names] p5.

include/ck/tensor_operation/gpu/block/blockwise_gemm_pipeline_xdlops_v1_ab_scale.hpp

@@ -308,7 +308,7 @@ struct BlockwiseGemmXdlops_pipeline_v1_ab_scale<BlockGemmPipelineScheduler::Intr
                                 typename vector_type<ComputeDataType,
                                                      xdlops_gemm.K1PerXdlops>::type;
 
-                            xdlops_gemm.template Run(
+                            xdlops_gemm.template Run<>(
                                 a_thread_vec.template AsType<mfma_input_type>(),
                                 b_thread_vec.template AsType<mfma_input_type>(),
                                 c_thread_buf_per_scale.GetVectorTypeReference(I0));
@@ -390,7 +390,8 @@ struct BlockwiseGemmXdlops_pipeline_v1_ab_scale<BlockGemmPipelineScheduler::Intr
                         using mfma_input_type =
                             typename vector_type<ComputeDataType, xdlops_gemm.K1PerXdlops>::type;
 
-                        xdlops_gemm.template Run(a_thread_vec.template AsType<mfma_input_type>(),
+                        xdlops_gemm.template Run<>(
+                            a_thread_vec.template AsType<mfma_input_type>(),
                             b_thread_vec.template AsType<mfma_input_type>(),
                             c_thread_buf_per_scale.GetVectorTypeReference(I0));
                     });

include/ck/tensor_operation/gpu/block/blockwise_gemm_pipeline_xdlops_v2_ab_scale.hpp

@@ -350,7 +350,7 @@ struct BlockwiseGemmXdlops_pipeline_v2_ab_scale<BlockGemmPipelineScheduler::Intr
                                     typename vector_type<ComputeDataType,
                                                          xdlops_gemm.K1PerXdlops>::type;
 
-                                xdlops_gemm.template Run(
+                                xdlops_gemm.template Run<>(
                                     a_thread_vec.template AsType<mfma_input_type>(),
                                     b_thread_vec.template AsType<mfma_input_type>(),
                                     c_thread_buf_per_scale.GetVectorTypeReference(I0));
@@ -443,7 +443,7 @@ struct BlockwiseGemmXdlops_pipeline_v2_ab_scale<BlockGemmPipelineScheduler::Intr
                                 typename vector_type<ComputeDataType,
                                                      xdlops_gemm.K1PerXdlops>::type;
 
-                            xdlops_gemm.template Run(
+                            xdlops_gemm.template Run<>(
                                 a_thread_vec.template AsType<mfma_input_type>(),
                                 b_thread_vec.template AsType<mfma_input_type>(),
                                 c_thread_buf_per_scale.GetVectorTypeReference(I0));
@@ -518,7 +518,8 @@ struct BlockwiseGemmXdlops_pipeline_v2_ab_scale<BlockGemmPipelineScheduler::Intr
                         using mfma_input_type =
                             typename vector_type<ComputeDataType, xdlops_gemm.K1PerXdlops>::type;
 
-                        xdlops_gemm.template Run(a_thread_vec.template AsType<mfma_input_type>(),
+                        xdlops_gemm.template Run<>(
+                            a_thread_vec.template AsType<mfma_input_type>(),
                             b_thread_vec.template AsType<mfma_input_type>(),
                             c_thread_buf_per_scale.GetVectorTypeReference(I0));
                     });
@@ -575,7 +576,8 @@ struct BlockwiseGemmXdlops_pipeline_v2_ab_scale<BlockGemmPipelineScheduler::Intr
                         using mfma_input_type =
                             typename vector_type<ComputeDataType, xdlops_gemm.K1PerXdlops>::type;
 
-                        xdlops_gemm.template Run(a_thread_vec.template AsType<mfma_input_type>(),
+                        xdlops_gemm.template Run<>(
+                            a_thread_vec.template AsType<mfma_input_type>(),
                             b_thread_vec.template AsType<mfma_input_type>(),
                             c_thread_buf_per_scale.GetVectorTypeReference(I0));
                     });

include/ck/tensor_operation/gpu/block/blockwise_gemm_pipeline_xdlops_v3_ab_scale.hpp

@@ -427,7 +427,7 @@ struct BlockwiseGemmXdlops_pipeline_v3_ab_scale<BlockGemmPipelineScheduler::Intr
                                 typename vector_type<ComputeDataType,
                                                      xdlops_gemm.K1PerXdlops>::type;
 
-                            xdlops_gemm.template Run(
+                            xdlops_gemm.template Run<>(
                                 a_thread_vec.template AsType<mfma_input_type>(),
                                 b_thread_vec.template AsType<mfma_input_type>(),
                                 c_thread_buf_per_scale.GetVectorTypeReference(I0));
@@ -504,7 +504,8 @@ struct BlockwiseGemmXdlops_pipeline_v3_ab_scale<BlockGemmPipelineScheduler::Intr
                         using mfma_input_type =
                             typename vector_type<ComputeDataType, xdlops_gemm.K1PerXdlops>::type;
 
-                        xdlops_gemm.template Run(a_thread_vec.template AsType<mfma_input_type>(),
+                        xdlops_gemm.template Run<>(
+                            a_thread_vec.template AsType<mfma_input_type>(),
                             b_thread_vec.template AsType<mfma_input_type>(),
                             c_thread_buf_per_scale.GetVectorTypeReference(I0));
                     });