Update on "[Executorch] Refactor op_mul's broadcasting utils"

Summary:
Refactoring broadcast handling utils that were added for op_mul. This is
in prepartion use these utils to handle broadcast for other ops such as
add, sub, div.
Plus remove a redundant test

Test Plan:
optimized_kernels_test in CI

Reviewers:

Subscribers:

Tasks:

Tags:

cc larryliu0820 manuelcandales

Differential Revision: [D69491816](https://our.internmc.facebook.com/intern/diff/D69491816)

[ghstack-poisoned]

Author: kimishpatel

kernels/optimized/cpu/binary_ops.h

@@ -199,7 +199,6 @@ Tensor& handle_last_dim_broadcast_elementwise(
     const Tensor& b,
     Tensor& out,
     const ElementwiseOptimizedPath selected_optimized_path) {
-  ScalarType out_type = out.scalar_type();
   const Tensor* lhs;
   const Tensor* rhs;
   if (selected_optimized_path ==