Move source file into sub-directories

library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_c_shuffle_bf16_bf16_bf16/CMakeLists.txt

+target_sources(${INSTANCE_LIBRARY_NAME} PRIVATE
+  device_gemm_xdl_c_shuffle_bf16_bf16_bf16_km_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_bf16_bf16_bf16_km_nk_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_bf16_bf16_bf16_mk_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_bf16_bf16_bf16_mk_nk_mn_instance.cpp
+)

library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_c_shuffle_f16_f16_f16/CMakeLists.txt

+target_sources(${INSTANCE_LIBRARY_NAME} PRIVATE
+  device_gemm_xdl_c_shuffle_2_stage_f16_f16_f16_mk_nk_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instance.cpp
+)

library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_c_shuffle_f32_f32_f32/CMakeLists.txt

+target_sources(${INSTANCE_LIBRARY_NAME} PRIVATE
+  device_gemm_xdl_c_shuffle_f32_f32_f32_km_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f32_f32_f32_km_nk_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f32_f32_f32_mk_kn_mn_instance.cpp
+  device_gemm_xdl_c_shuffle_f32_f32_f32_mk_nk_mn_instance.cpp
+)