test: skip failing torch-xpu-ops UT cases in P0 scope

[libohao1201]

Summary

Skip 110 failing P0-scope UT cases by injecting unittest.skip on the exact generated test methods after instantiation — no change to skip_list_common.py.

Each skip carries its tracking issue link, so the skip reason points directly at the owning issue.

Approach

For each test file, after instantiate_device_type_tests(...) (or instantiate_parametrized_tests(...) for XCCL), a small helper iterates a class -> {exact_method_name: issue_url} map and applies unittest.skip(reason) to the matching generated methods. The helper is wrapped in a function so no TestCase-bound loop variable leaks into module globals (which would otherwise cause pytest to re-collect the class).

Files changed

File	Cases skipped
test/xpu/test_ops_xpu.py	61
test/xpu/test_unary_ufuncs_xpu.py	31
test/xpu/test_optim_xpu.py	11
test/xpu/distributed/test_c10d_xccl.py	6
test/xpu/test_shape_ops_xpu.py	1
Total	110

Validation

• All 5 files compile (py_compile).
• End-to-end pytest --collect-only / run confirms the targeted methods are collected exactly once and reported SKIPPED with the issue link as reason.

[libohao1201]

@copilot https://gh.io/copilot-coding-agent-docs https://gh.io/copilot-coding-agent-docs https://gh.io/copilot-coding-agent-docs The Python lint (flake8/ruff) check failed. Please fix the lint errors in this PR.

Important: Do NOT update the plan in the PR description directly. Use reply comments to update the status.

Lint job log: https://github.com/intel/torch-xpu-ops/actions/runs/32924365796

Instructions:

Review the lint failure log via the link above
Fix all lint errors in the relevant source files
Ensure both Python lint and Clang format checks pass
Use reply comments to report progress - do NOT modify the PR description plan directly

[Copilot]

@copilot https://gh.io/copilot-coding-agent-docs https://gh.io/copilot-coding-agent-docs https://gh.io/copilot-coding-agent-docs The Python lint (flake8/ruff) check failed. Please fix the lint errors...

Addressed in 2e7c190. I fixed the PYFMT issues reported by run 32924365796 by adding the formatter-required blank lines before the new helper functions and correcting the import order in test_optim_xpu.py. Local validation: python -m py_compile on the 5 touched test files, git diff --check, secret scan, and automated code review all passed.

[mengfei25]

Could we skip them here as well, similar to how they are handled in skip_list_common.py? or another file, similar as other skips?
Maintaining two different skipping mechanisms might make the code harder to maintain down the road.
What do you think?