Implement aten::nonzero_static on XPU backend

[Silv3S]

Fixes #1943

[Copilot]

Pull Request Overview

This PR implements the nonzero_static operation for the XPU backend, adding support for a static-sized version of the nonzero operation that fills output to a specified size. This addresses issue #1943 in the torch-xpu-ops repository.

• Adds dispatch entries for nonzero_static and nonzero_static.out operations in the native functions YAML
• Implements the XPU backend functions with proper input validation and size handling
• Refactors common validation logic into a shared helper function

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
yaml/native/native_functions.yaml	Adds dispatch entries for nonzero_static operations to XPU backend
src/ATen/native/xpu/Nonzero.cpp	Implements nonzero_static functions and refactors common validation logic

---

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

[EikanWang]

The logic looks good to me. Pls. help add test cases or list the existing test cases here as evidence that validates the logic.

[Silv3S]

There are two parametrized tests from upstream, aligned to work with XPU

• test_unary_ufuncs_xpu.py::TestUnaryUfuncsXPU::test_nonzero_static_large_xpu
• test_unary_ufuncs_xpu.py::TestUnaryUfuncsXPU::test_nonzero_static_xpu

[CuiYifeng]

LGTM.