Replace use of deprecated sub group size attribute with SYCL 2020 version

[jbrodman]

The attribute [[intel::reqd_sub_group_size]] was deprecated years ago in favor of the standardized version [[sycl::reqd_sub_group_size]].

[Copilot]

Pull Request Overview

This PR updates deprecated Intel-specific SYCL attributes to use the standardized SYCL 2020 version across multiple kernel implementations. The change replaces [[intel::reqd_sub_group_size]] with [[sycl::reqd_sub_group_size]] to align with modern SYCL standards.

• Replace deprecated Intel-specific sub group size attributes with SYCL 2020 standard
• Update kernel functors across various XPU SYCL implementations
• Maintain existing sub group size values while modernizing the attribute syntax

Reviewed Changes

Copilot reviewed 13 out of 13 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
TensorModeKernel.cpp	Updated ComputeModeKernelFunctor attribute
SortingKernels.h	Updated multiple sorting kernel functor attributes
SoftMaxKernels.cpp	Updated forward and backward softmax kernel attributes
Norm.h	Updated FusedNormKernelFunctor and RowwiseMomentsKernelFunctor attributes
MultiLabelMarginLossKernels.cpp	Updated loss kernel functor attributes
LossNLL2dKernels.cpp	Updated NLL loss kernel functor attributes
LinearInt4.cpp	Updated LinearInt4KernelFunctor attribute
LayerNormKernels.cpp	Updated layer norm kernel functor attributes
IndexKernelUtils.h	Updated VectorizedGatherKernel attribute
GroupNormKernels.cpp	Updated multiple group norm kernel functor attributes
ForeachReduceKernels.cpp	Updated foreach reduce kernel functor attributes
Dequant_int4.cpp	Updated DequantInt4KernelFunctor attribute
BatchNormKernels.cpp	Updated multiple batch norm kernel functor attributes

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}