Brain_Data constructor flattens multiple files instead of preserving subject structure

[ssadhukha]

When loading multiple nifti files using the constructor Brain_Data(file_list), the resulting object flattens the data into a single vector instead of maintaining the multi-subject structure, which is making it difficult to perform group-level stuff.

Workaround I found:

all_subjects = Brain_Data()
for file in rsa_files:
    subject = Brain_Data(file)
    if all_subjects.isempty():
        all_subjects = subject
    else:
        all_subjects = all_subjects.append(subject)

This worked previously? I was always able to just do all_subjects = Brain_Data(rsa_files) and the multi-subject structure was preserved.

python version: 3.10.18
nltools version: 0.5.1