Improve TOC generation #11
Description
Right now, TOC scans every published file on every change which is kinda expensive. This should probably be changed to dynamo so we can do fragment updates to the TOC in the DB avoiding race conditions. This will make it so we don't have to scan all the files, just the one that was changed/updated. After the dynamo object is updated, an S3 copy should be made in the same place and the same format.