Fix include paths in profiler and tests.

profiler/include/profile_batched_gemm_masking_scale_softmax_gemm_permute_impl.hpp

@@ -7,7 +7,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
-#include "ck/tensor_operation/gpu/device/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/tensor_operation_instance/gpu/batched_gemm_masking_scale_softmax_gemm_permute.hpp"

test/batched_gemm_masking_scale_softmax_gemm_permute/test_batched_gemm_masking_scale_softmax_gemm_permute_util.hpp

@@ -5,7 +5,7 @@
 
 #include <vector>
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp"
 #include "profiler/include/profile_batched_gemm_masking_scale_softmax_gemm_permute_impl.hpp"
 using ck::tensor_operation::device::GemmSpecialization;