babs requires all zips in original input dataset to be downloaded before doing babs-init

[audreycluo]

Summary

Instead of datalad getting each subject as babs runs the BIDS app, babs requires that all zips have been gotten in the input BIDS dataset before doing babs-init. These zips need to remain 'gotten' throughout each step of babs. If I drop zips in my original BIDS dataset (usually a datalad clone) after babs-init, babs fails to run the BIDS app.

It'd be great if we don't have to download all the zips in the original datalad clone because it takes up a lot of space and takes time.