merge from public repo

example/62_convnd_activ/convscale_relu/CMakeLists.txt

+list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
+set(target 0)
+foreach(gpu IN LISTS GPU_TARGETS)
+ if(gpu IN_LIST gpu_list AND target EQUAL 0)
+     add_custom_target(example_convnd_activ_xdl_convscale_relu)
+     add_example_executable(example_convnd_fwd_xdl_convscale_relu_fp8 convnd_fwd_xdl_convscale_relu_fp8.cpp)
+     add_example_dependencies(example_convnd_activ_xdl_convscale_relu example_convnd_fwd_xdl_convscale_relu_fp8 )
+
+     set(target 1)
+ endif()
+endforeach()

example/65_gemm_multiply_multiply/CMakeLists.txt

-add_example_executable(example_gemm_multiply_multiply_xdl_fp16 gemm_multiply_multiply_xdl_fp16.cpp)
+add_example_executable(example_gemm_multiply_multiply_xdl_fp8 gemm_multiply_multiply_xdl_fp8.cpp)
+add_example_executable(example_gemm_multiply_multiply_xdl_fp8_ab_scale gemm_multiply_multiply_xdl_fp8_ab_scale.cpp)
+add_example_executable(example_gemm_add_add_xdl_fp16 gemm_add_add_xdl_fp16.cpp)