feat(ci): run CI pipeline on ghcr.io containers, not cvmfs

[wdconinc]

Briefly, what does this PR introduce?

Instead of using the cvmfs container (with inherent lag) this switches some of the CI over to using the containers on ghcr.io directly. We'll assess how well/fast this works.

What kind of change does this PR introduce?

• Bug fix (issue #__)
• New feature (issue #__)
• Documentation update
• Other: __

Please check if this PR fulfills the following:

• Tests for the changes have been added
• Documentation has been added / updated
• Changes have been communicated to collaborators

Does this PR introduce breaking changes? What changes might users need to make to their code?

No.

Does this PR change default behavior?

No.

[wdconinc]

• 33s overhead for setting up cvmfs job (some overhead to be expected on first use later on), 1m53s for clang build only,
• 3m02s overhead for pulling a full container (no additional overhead expected), 1m30s for clang build only.

All this indicates that sticking to cvmfs for CI (total overhead 56s) continues to have benefits over containers (total overhead 3m02s) until a factor ~3 decrease in container size. This is unlikely to be possible due to the irreducible large components in the form of geant4 data files.