Build: Expand CUDA Toolkit Matrix

[matthewdouglas]

This PR changes the build action to build with a wider matrix of CUDA Toolkit versions, in preparation for a release. The COMPUTE_CAPABILITY build option is set appropriately based on the support from each toolkit version.

CUDA	Targets
11.7.1	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, compute_86
11.8.0	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, sm_89, compute_89
12.0.1	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, sm_89, sm_90, compute_90
12.1.1	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, sm_89, sm_90, compute_90
12.2.2	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, sm_89, sm_90, compute_90
12.3.2	sm_50, sm_52, sm_60, sm_61, sm_70, sm_75, sm_80, sm_86, sm_89, sm_90, compute_90

[Titus-von-Koeller]

Thanks @matthewdouglas to create this PR for us to collaborate on, based on what we spoke about in Slack :)

[github-actions]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.