Explicitly define volumetric space and resolution in CIFTI outputs

[tsalo]

What would you like to see added in fMRIPrep?

Given that fsLR only describes the surface space, and the volume space of current CIFTI outputs (i.e., MNI152NLin6Asym) is only implied by convention, I think we should add volspace-MNI152NLin6Asym to fsLR-space CIFTI outputs. The volspace entity was proposed in BEP011 Structural Derivatives.

I also was thinking about the density and resolution entities. Does it make sense to use 91k as shorthand for den-32k + res-2, or should we change that?

Do you have any interest in helping implement the feature?

Yes

Additional information / screenshots

This stems from #3330.
This will require changes to sMRIPrep, Nibabies, and ASLPrep at minimum.

[tsalo]

I imagine that, since this would change output filenames, it would require a minor release. Might even need to wait until 25.0.0?

[tsalo]

I've opened bids-standard/bids-specification#1900 to add volspace to BIDS.

[tsalo]

I've closed that BIDS PR in favor of using +s (@effigies' idea), so I'm going to close this issue. I can open new issues if relevant changes need to be made to fMRIPrep.