COMPMID-2072: Use a constexpr for num_elems_processed_per_iteration where possible

Change-Id: I26cb699ae3a77003ef7d05ac30d3ed518214e25f
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1375
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Manuel Bottini <manuel.bottini@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
diff --git a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
index e878dbc..0c0b0ec 100644
--- a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp
@@ -38,12 +38,9 @@
 #include <cstddef>
 #include <cstdint>
 
-using namespace arm_compute;
-
 namespace arm_compute
 {
 class Coordinates;
-} // namespace arm_compute
 
 namespace
 {
@@ -56,7 +53,7 @@
 }
 std::pair<Status, Window> validate_and_configure_window_matrix_a_reduction(ITensorInfo *input, ITensorInfo *output)
 {
-    const unsigned int num_elems_processed_per_iteration = 1;
+    constexpr unsigned int num_elems_processed_per_iteration = 1;
 
     Window win = calculate_max_window(*output, Steps(num_elems_processed_per_iteration));
 
@@ -221,3 +218,4 @@
     }
     while(collapsed.slide_window_slice_2D(slice_out));
 }
+} // namespace arm_compute