New ability to read in DICOMs for the proton scan

[davidmummy]

Current functionality is to recon a UTE scan from a Twix or MRD file.

However, for certain sites (and just for overall flexibility) it will be useful to be able to read in a proton scan from DICOM files and use that instead. As the proton scan is not strictly necessary for segmentation and not at all necessary for quantification, we have some flexibility here as to the nature of the input images.

• This is not necessarily a UTE scan, so we can be agnostic to the acquisition type -- just that it's a proton scan.
• I believe all we need to read in as far as meta-data are the voxel dimensions. The ANTs registration should take care of everything else. There is example code in the v3 pipeline that does this already.
• This will be an additional option in the config file which when a DICOM proton path is specified will bypass the normal UTE recon and just continue the rest of the processing as normal using the image pulled from the DICOM (and then turned into a nifti, etc). Essentially just "injecting" the dicom-based images at the appropriate point in the pipeline, and after that it doesn't really matter where it came from.

Assigning to @annacostelle with help from @sl609 and @junlanlu on how best to implement.

[davidmummy]

Closing by previous commit from a little earlier we are still learning this