Update on "[ExecuTorch] Add broadcasting support to optimized op_div"

Summary:
Similar to broadcast support in op_mul

Test Plan:
Tests added

Reviewers:

Subscribers:

Tasks:

Tags:

cc larryliu0820 manuelcandales

[ghstack-poisoned]

Author: kimishpatel

kernels/optimized/cpu/op_div.cpp

@@ -137,15 +137,13 @@ Tensor& opt_div_out(
       // to be agnostic of op. This should be fixed, likely by moving lambda
       // creation to handle_broadcast_elementwise and it be aware of which op is
       // being executed.
-      auto div_lambda = [](auto x, auto y, auto alpha) {
-        [[maybe_unused]] alpha;
+      auto div_lambda = [](auto x, auto y, [[maybe_unused]] auto alpha) {
         return y / x;
       };
       return torch::executor::handle_broadcast_elementwise<op_name>(
           ctx, div_lambda, a, b, out, selected_optimized_path);
     } else {
-      auto div_lambda = [](auto x, auto y, auto alpha) {
-        [[maybe_unused]] alpha;
+      auto div_lambda = [](auto x, auto y, [[maybe_unused]] auto alpha) {
         return x / y;
       };
       return torch::executor::handle_broadcast_elementwise<op_name>(

kernels/optimized/cpu/op_mul.cpp

@@ -133,8 +133,7 @@ Tensor& opt_mul_out(
     // Reason for using alpha even when used for mul is becasuse
     // handle_broadcast_elementwise is used for add and sub as well
     // and it uses alpha.
-    auto mul_lambda = [](auto x, auto y, auto alpha) {
-      [[maybe_unused]] alpha;
+    auto mul_lambda = [](auto x, auto y, [[maybe_unused]] auto alpha) {
       return x * y;
     };
     static constexpr const char op_name[] = "mul.out";