pipeline update

[mkutu]

happens on a per batch basis and live in the annotation pipeline

While we're generating cutouts, we create the metadata (.json) files and also push to the database.

This is a couple lines of code within the larger annotation pipeline framework that might pull from the same script and config as the bulk update #168