Noisy and useless summary when specifying a list of subjects manually

[ghisvail]

Describe the bug I get a summary of the subjects not found in the BIDS folder despite manually specifying a list of subjects via the -sl / --subjects-list option. If there is anything useful to report in this case, it's the absence of specific subjects from the manually specified list in the target BIDS folder, instead of the whole list found in the clinical data. It yields an unecessary amount of noise to the end of the logs, especially when the dataset features a lot of participants such as ADNI.

To Reproduce Steps to reproduce the behavior:

• echo $SUBJECT_ID > /tmp/participants.txt, for instance 021_S_0178.
• clinica convert adni-to-bids --subjects-list /tmp/participants.txt /path/to/imaging_data /path/to/clinical_data /path/to/bids_data

Expected behavior If the desired $SUBJECT_ID is found in the target BIDS folder, then nothing. If the desired $SUBJECT_ID is not found in the target BIDS, then report it.

[AliceJoubert]

Still the case for ADNI and OASIS datasets since both use create_participants_df from bids_utils. To discuss :

• [ ] Should that list be straight up the diff between desired subjects (from subj list or raw dir) the bids output or only subjects that were not identified as problematic (ex : invalid field mapping case) ?
• [ ] Might interact with issue #1189