[ENH] Proposition for adni-to-bids converter FMAP renaming

[AliceJoubert]

Refactor of adni-to-bids FMAP modality renaming functionality to address part of issue #1174

It is based on the outputs of dcm2nix (https://github.com/rordenlab/dcm2niix/blob/master/FILENAMING.md) and what is expected in BIDS specifications (https://bids-specification.readthedocs.io/en/stable/modality-specific-files/magnetic-resonance-imaging-data.html#types-of-fieldmaps).

The recognition of cases is based on file extension (e1, e1_ph...) and the number of files present in the directory. The files are renamed only if a BIDS case is recognized. Else, the files are deleted. Case 1 and 2 are handled, not cases 3 and 4.

For BIDS case 3 (GE scanners) :

• not in ADNI https://adni.loni.usc.edu/support/experts-knowledge-base/question/?QID=1027 ie no need to support it
• confusing case https://neurostars.org/t/does-direct-field-mapping-for-field-map-require-magnitude-nii-gz/20299/6

Should be handled in other PR :

• Bids Guess option : integrating what was suggested in PR [ENH] Proposition for adni_fmap renaming function #1175
• After conversion deletion and checks - continuation from PR [FIX] Properly deleting real and imaginary files for FMAP in adni-to-bids converter. #1188

[AliceJoubert]

Logic could be greatly simplified if we decided to only support case 1 and case 2 from BIDS. Since Case 3 does not seem to be represented in ADNI.

[NicolasGensollen]

Thanks for taking care of this @AliceJoubert !
I made a quick first pass with a few suggestions along the way.
The main suggestion will probably change a few things so I'll make a second round once implemented if you agree with it.
It simply restructures a bit what you implemented using slightly more robust patterns.

[NicolasGensollen]

Thanks @AliceJoubert !
A small suggestion to push what you did one step further to avoid the duplication of the renaming logic.

[NicolasGensollen]

Thanks @AliceJoubert !
A few other small suggestions to make the code a bit DRYer.

[NicolasGensollen]

Thanks @AliceJoubert !
Another small suggestion to remove additional redundancy.

[NicolasGensollen]

Thanks @AliceJoubert for taking my comments into account !
I believe the code looks good. I just made very minor suggestions to ease readability.
With a few unit tests this should be good to go ! 🚀

[AliceJoubert]

Thanks @NicolasGensollen for the comments ! I have added some unit tests, do you mind looking at them ?

[NicolasGensollen]

Thanks @AliceJoubert !

I made one or two suggestions but I think the tests look great ! 👍
Could you also rebase on latest dev ? The CI seems not happy for some reason.

[AliceJoubert]

Thanks for the review @NicolasGensollen ! I modified the tests according to your suggestions. Do you see anything else ?

[NicolasGensollen]

LGTM, thanks @AliceJoubert !