Overhaul PyDesigner Code 2

[TheJaeger]

This PR fixes various reported bugs and updates documentation to tackle:

• Error in determining maximum BVAL #303
• motion plot fails after eddy_quad generates an error #305
• The meaning of DTI parameters #306
• High value voxels #308
• Transition to Poetry
• Add type hints
• Overhaul docstring
• Introduce pre-commit hooks
• Add pytests
• Remove redundant classes and functions

In addition, PyDesigner will undergo transition to Poetry for package management and pre-commit for hooks and tests. Automations will also be introduced to automatically publish, release, and document changes.

Another big change is to transition PyDesigner into a diffusion imaging library i.e. various functions of PyDesigner can be called for custom scripting.

Closes #303 #305 #306 #308

[TheJaeger]

@ryn-thorn @gonzoBlackMamba This update is ready. Please test on your end and let me know if it works fine. You need to have Poetry installed and need AT LEAST Python 3.12.

Switch to this branch and run poetry install. Then invoke pydesigner with poerty run pydesigner [OPTIONS] .... Test some data and approve/reject this pull request accordingly.

[github-actions]

Package	Line Rate	Health
.	0%	❌
fitting	41%	➖
plotting	3%	❌
postprocessing	0%	❌
preprocessing	23%	➖
system	83%	✔
tractography	16%	➖
Summary	25% (945 / 3786)	➖

Minimum allowed line rate is 10%

[gonzoBlackMamba]

@TheJaeger this is great! I got it to run on a test data set. Two things immediately come to my attention:

1. The B0.nii, B1000.nii, B2000.nii and B*000.nii images are empty (zero-valued everywhere), and

2. the fbi_odf_raw.nii files matches with MRTrix3 just fine:

However, the fbi_odf_mrtrix.nii file is flipped in the axial and coronal directions (see images)

[gonzoBlackMamba]

@TheJaeger I was also under the impression that we needed everything in RAS, however the dwi_raw and dwi_preprocessed images are LPS:

[gonzoBlackMamba]

@TheJaeger: as are the output metric image maps (for instance the FAA)

[gonzoBlackMamba]

@TheJaeger for the fbi_odf_mrtrix.nii file, not certain, but you may need to flip the gradient directions in the x- and y-directions (the ones that make up the azimuthal angle ('phi') to match the RAS orientation of MRtrix...

[gonzoBlackMamba]

@TheJaeger last thing for now: I see that on the command line process the GFA and peak detection register 0% while everything else completes normally