Build with CUDA 12.6

[jeongseok-meta]

This is a temporary solution to support CUDA 12.6 until the issue at conda-forge/conda-forge-pinning-feedstock#6630 is resolved. The zipped configs are intended to add an additional build for 12.6 while keeping the existing builds for None, 11.8, and 12.0 untouched. Once the global pinning for the CUDA version is set, the manual zipped configs can be removed

Checklist

• Used a personal fork of the feedstock to propose changes
• Bumped the build number (if the version is unchanged)
• Reset the build number to 0 (if the version changed)
• Re-rendered with the latest conda-smithy (Use the phrase @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• Ensured the license file is being packaged.

[jeongseok-meta]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you, but it looks like there was nothing to do.

_{This message was generated by GitHub actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/11691661628.}

[weiji14]

Should the CUDA versions be set manually per feedstock rather than in the global pinnings? I'm reading the thread at conda-forge/conda-forge-pinning-feedstock#6630, and it seems like there's no consensus yet on whether to have 3 versions (11.8, 12.0, 12.6) or just 2 versions (11.8, 12.6).

[jeongseok-meta]

Yes, you're right, but this change supports current pinning versions (11.8 and 12.0) and adds 12.6, as pytorch has done. It's a stopgap solution to unblock users until a consensus is reached. I kindly suggest adding 12.6 now, but ultimately it's up to the maintainer.

[weiji14]

I have a slight preference to rely on conda-forge's global-pinnings so that we don't need to unpin things in this recipe/conda_build_config.yaml file in the future (e.g. when a newer CUDA version comes out). But no strong opinions either on adding CUDA 12.6 here, especially if any users require it right now (unsure if CUDA 12.6 unlocks any new functionality in pytorch3d or dependent libraries). Any thoughts @Tobias-Fischer?

[Tobias-Fischer]

Happy to merge here @jeongseok-meta - if you could please revert the manual changes in the conda_build_config.yaml once there is a conda-forge agreed migration that'd be amazing :).