Full and incremental backups
let me try to explain how it works.
A full backup is usually considered as a set of all files on source falling under the selected criteria.
In fact, what we do for file backup (and this is what's available for Linux currently), is incremental and block-level incremental.
Unlike regular schedule for block-level, where we send deltas of the files, with Full backup schedule, we back up full versions of those files (entire files), but it's actually incremental.
To create the real full (consisting of all files) you need to change the destination folder or a bucket and it will perform the real full, but I realize it's too much of overhead.
I will file an enhancement request to fix the wording.