Enable apt proxy caching; skip hosted Windows builds

[kkraus14]

Description

closes N/A

• Enable apt proxy caching for proxy-cache steps.
• Skip proxy cache on GitHub-hosted Windows CPU builds.
• Update SPDX year headers touched by pre-commit.

Checklist

• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[copy-pr-bot]

Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[rwgk]

Nice! I noticed before that downloading Windows wheels from PyPI (for pathfinder testing) was very slow (maybe even rate limiting).

[kkraus14]

/ok to test

[jrcichra]

You'll need to exclude the GitHub hosted runners from using that action. It only works on our self-hosted runners.

[kkraus14]

For some reason I had thought that we moved our Windows CPU builds to the CPU pool. I'll address it, thanks.

[kkraus14]

/ok to test

[kkraus14]

Nice! I noticed before that downloading Windows wheels from PyPI (for pathfinder testing) was very slow (maybe even rate limiting).

Unfortunately it failed because we're still on GitHub hosted Windows CPU runners. Hopefully we can move to self hosted CPU runners sooner than later.

[github-actions]

Doc Preview CI
Preview removed because the pull request was closed or merged.

[rwgk]

Nice! I noticed before that downloading Windows wheels from PyPI (for pathfinder testing) was very slow (maybe even rate limiting).

Unfortunately it failed because we're still on GitHub hosted Windows CPU runners. Hopefully we can move to self hosted CPU runners sooner than later.

I think it's still a big gain. This is the slow step I had in mind, in .github/workflows/test-wheel-windows.yml:

          pip install --only-binary=:all: -v ./*.whl --group "test-cu${TEST_CUDA_MAJOR}"

If I'm doing the math correctly, that's a whopping 2.3 GB; see below; I didn't actually realize before it's that much.

Luckily that's in the file where we now have enable-apt: true.

---

This is what that pip install command pulls in:

Using cached cuda_toolkit-13.1.1-py2.py3-none-any.whl (2.4 kB)
Using cached nvidia_cublas-13.2.1.1-py3-none-manylinux_2_27_x86_64.whl (401.1 MB)
Using cached nvidia_cuda_cccl-13.1.115-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (3.4 MB)
Using cached nvidia_cuda_nvcc-13.1.115-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (42.3 MB)
Using cached nvidia_cuda_nvrtc-13.1.115-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (46.6 MB)
Using cached nvidia_cuda_runtime-13.1.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (2.3 MB)
Using cached nvidia_cufft-12.1.0.78-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (226.1 MB)
Using cached nvidia_cufile-1.16.1.26-py3-none-manylinux_2_27_x86_64.whl (1.3 MB)
Using cached nvidia_curand-10.4.1.81-py3-none-manylinux_2_27_x86_64.whl (60.0 MB)
Using cached nvidia_cusolver-12.0.9.81-py3-none-manylinux_2_27_x86_64.whl (202.0 MB)
Using cached nvidia_cusparse-12.7.3.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (153.2 MB)
Using cached nvidia_npp-13.0.3.3-py3-none-manylinux_2_27_x86_64.whl (135.6 MB)
Using cached nvidia_nvfatbin-13.1.115-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (979 kB)
Using cached nvidia_nvjitlink-13.1.115-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (40.9 MB)
Using cached nvidia_nvjpeg-13.0.3.75-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (3.7 MB)
Using cached nvidia_nvvm-13.1.115-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (63.4 MB)
Using cached pytest-9.0.2-py3-none-any.whl (374 kB)
Using cached pluggy-1.6.0-py3-none-any.whl (20 kB)
Using cached cutensor_cu13-2.5.0-py3-none-manylinux2014_x86_64.whl (208.5 MB)
Using cached nvidia_cublasmp_cu13-0.7.0.125-py3-none-manylinux_2_28_x86_64.whl (6.6 MB)
Using cached nvidia_cudss_cu13-0.7.1.6-py3-none-manylinux_2_17_x86_64.whl (65.0 MB)
Using cached nvidia_cufftmp_cu13-12.1.3.2-py3-none-manylinux_2_28_x86_64.whl (205.6 MB)
Using cached nvidia_cusparselt_cu13-0.8.1-py3-none-manylinux2014_x86_64.whl (170.1 MB)
Using cached nvidia_nccl_cu13-2.29.2-py3-none-manylinux_2_18_x86_64.whl (201.1 MB)
Using cached nvidia_nvshmem_cu13-3.5.19-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (72.4 MB)
Using cached iniconfig-2.3.0-py3-none-any.whl (7.5 kB)
Using cached packaging-26.0-py3-none-any.whl (74 kB)
Using cached pygments-2.19.2-py3-none-any.whl (1.2 MB)
Using cached nvidia_cuda_crt-13.1.115-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (132 kB)