[XPU] Upgrade the XPU support packages version to 2025.1 in CI/CD

[chuanqi129]

As the XPU support packages deep learning essential 2025.1 has been public released. We target use 2025.1.1 for PyTorch 2.8 finally, but plan to upgrade to 2025.1.0 early to mitigate the upgrade risk.

• Dependencies

• • Land [XPTI] introduce new interface to kineto for PTI-0.11 and PTI-0.12 kineto#1066 for PTI 0.11.0 new interface in 2025.1.0
• • Update Kineto submodule version to include this change [Kineto] Upgrade the kineto commit to fb36cce #152007
• • Enable XCCL build in CI by [CI] Enable XCCL in XPU CI build #150927
• • Enable oneMKL build for XPU in CI

• Step 1. Upgrade 2025.1.0

• • Upgrade 2025.1.0 CI test both for Linux & Windows and add new docker image pytorch-linux-jammy-xpu-2025.1-py3 for Linux [BE] Upgrade XPU support package to 2025.1 in CICD #151899
• • Upgrade CD build with 2025.1.0 and add new runtime Pypi packages dependencies both for Linux & Windows CD whls [BE] Upgrade XPU support package to 2025.1 in CICD #151899
• • pytorch/test-infra repo update [BE] Upgrade XPU support package to 2025.1 in CICD test-infra#6553
• • pytorch libs repo update if need

• Step 2. Upgrade 2025.1.1

• • Upgrade 2025.1.1 CI test both for Linux & Windows
• • Upgrade CD build with 2025.1.1 and update the runtime pypi dependencies
• • pytorch/test-infra repo update if need
• • pytorch libs repo update if need

cc @gujinghui @EikanWang @fengyuan14 @guangyey