DWI unfolding

[jordandekraker]

The goal of this branch is to incorporate and simplify the code from https://github.com/uhussai7/HippoDiffusion.

Note that many steps are redundant since HippoDiffusion was built off an earlier version of HippUnfolding. We no longer need to worry about cropping, and code is already present to do things like interpolate between native and unfolded space, or reparamterize unfolded space according to real-world distances.

I will post an example script which demonstrates these, and will include an example subject that we can work off of

[akhanf]

Thanks for initiating this Jordan!

I would suggest the main interface to the diffusion unfolding be something like this:
function unfold_diffusion( in_hippunfold_results_dir, in_dwi_nii, in_dwi_bvec, in_dwi_bval, in_dwi_graddev, out_unfolddwi_dir )

where in_dwi_nii is the pre-processed diffusion-weighted 4d volume in the same space as the T2w (e.g. same as T1w in case of pre-processed HCP data)