merge changes from gpu_div branch

src/targets/gpu/CMakeLists.txt

@@ -40,8 +40,8 @@ add_library(migraphx_device
     device/div.cpp
     device/clip.cpp
     device/reduce_sum.cpp
-    device/pow.cpp
     device/reduce_mean.cpp
+    device/pow.cpp
 )
 set_target_properties(migraphx_device PROPERTIES EXPORT_NAME device)
 rocm_clang_tidy_check(migraphx_device)

src/targets/gpu/lowering.cpp

@@ -52,8 +52,8 @@
 #include <migraphx/gpu/convert.hpp>
 #include <migraphx/gpu/clip.hpp>
 #include <migraphx/gpu/reduce_sum.hpp>
-#include <migraphx/gpu/pow.hpp>
 #include <migraphx/gpu/reduce_mean.hpp>
+#include <migraphx/gpu/pow.hpp>
 #include <utility>
 #include <functional>
 #include <algorithm>