Fix review comments

Author: SergeyKopienko

include/oneapi/dpl/internal/scan_by_segment_impl.h

@@ -117,7 +117,7 @@ struct __sycl_scan_by_segment_impl
         // We should avoid using _ExecutionPolicy in __kernel_name_generator template params
         // because we always specialize this operator() calls only by _ExecutionPolicy as "const reference".
         // So, from this template param point of view, only one specialization is possible per concrete _ExecutionPolicy type.
-        // _ExecutionPolicy type information is encoded in _CustomName to distinguish between concrete policy types.
+        // _ExecutionPolicy type information is embedded in _CustomName to distinguish between concrete policy types.
         using _SegScanWgKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<
             _SegScanWgPhase, _CustomName, _Range1, _Range2, _Range3, _BinaryPredicate, _BinaryOperator>;
         using _SegScanPrefixKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<

include/oneapi/dpl/pstl/hetero/dpcpp/parallel_backend_sycl.h

@@ -242,7 +242,8 @@ struct __parallel_scan_submitter<_CustomName, __internal::__optional_kernel_name
 
         // We should avoid using _ExecutionPolicy in __kernel_name_generator template params
         // because we always specialize this operator() calls only by _ExecutionPolicy as "const reference".
-        // So, from this template param point of view, only one specialization is possible.
+        // So, from this template param point of view, only one specialization is possible per concrete _ExecutionPolicy type.
+        // _ExecutionPolicy type information is embedded in _CustomName to distinguish between concrete policy types.
         using _LocalScanKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<
             __scan_local_kernel, _CustomName, _Range1, _Range2, _Type, _LocalScan, _GroupScan, _GlobalScan>;
         using _GroupScanKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<

include/oneapi/dpl/pstl/hetero/dpcpp/parallel_backend_sycl_radix_sort.h

@@ -655,7 +655,8 @@ struct __parallel_radix_sort_iteration
 
         // We should avoid using _ExecutionPolicy in __kernel_name_generator template params
         // because we always specialize this submit() calls only by _ExecutionPolicy as "const reference".
-        // So, from this template param point of view, only one specialization is possible.
+        // So, from this template param point of view, only one specialization is possible per concrete _ExecutionPolicy type.
+        // _ExecutionPolicy type information is embedded in _CustomName to distinguish between concrete policy types.
         using _RadixCountKernel =
             __internal::__kernel_name_generator<__count_phase, _CustomName, std::decay_t<_InRange>,
                                                 std::decay_t<_TmpBuf>, _Proj>;

include/oneapi/dpl/pstl/hetero/dpcpp/parallel_backend_sycl_reduce.h

@@ -322,7 +322,8 @@ struct __parallel_transform_reduce_impl
 
         // We should avoid using _ExecutionPolicy in __kernel_name_generator template params
         // because we always specialize this submit() calls only by _ExecutionPolicy as "const reference".
-        // So, from this template param point of view, only one specialization is possible.
+        // So, from this template param point of view, only one specialization is possible per concrete _ExecutionPolicy type.
+        // _ExecutionPolicy type information is embedded in _CustomName to distinguish between concrete policy types.
         using _ReduceKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<
             __reduce_kernel, _CustomName, _ReduceOp, _TransformOp, _NoOpFunctor, _Ranges...>;
 

include/oneapi/dpl/pstl/hetero/dpcpp/parallel_backend_sycl_reduce_by_segment.h

@@ -98,8 +98,9 @@ __parallel_reduce_by_segment_fallback(oneapi::dpl::__internal::__device_backend_
     using _CustomName = oneapi::dpl::__internal::__policy_kernel_name<_ExecutionPolicy>;
 
     // We should avoid using _ExecutionPolicy in __kernel_name_generator template params
-    // because we always specialize this __parallel_reduce_by_segment_fallback calls only by _ExecutionPolicy as "const reference".
-    // So, from this template param point of view, only one specialization is possible.
+    // because we always specialize this __parallel_reduce_by_segment_fallback() calls only by _ExecutionPolicy as "const reference".
+    // So, from this template param point of view, only one specialization is possible per concrete _ExecutionPolicy type.
+    // _ExecutionPolicy type information is embedded in _CustomName to distinguish between concrete policy types.
     using _SegReduceCountKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<
         _SegReduceCountPhase, _CustomName, _Range1, _Range2, _Range3, _Range4, _BinaryPredicate, _BinaryOperator>;
     using _SegReduceOffsetKernel = oneapi::dpl::__par_backend_hetero::__internal::__kernel_name_generator<