Cleanup of the files in dataset storage that are not linked to that dataset #9130
Description
Overview of the Feature Request
As described by @qqmyers in #9002 comments:
W.r.t. temp files - we have this issue in general (i.e. if a user does a UI direct upload and leaves the page without hitting cancel or save, Dataverse doesn't know and doesn't clean up the files. Similarly in the direct upload API, if the final /addFiles call isn't done, the files are abandoned. In S3 systems that support it, we add a tag so these temp files can be easily found and deleted (i.e. via a cron job), but it would be nicer if there were an API call to remove files unrelated to a dataset on S3 (or file system, etc.) Such an API would have to deal with files in different versions and all of the auxiliary files, thumbnails, metadata exports that should remain but delete any files with storageidentifiers that aren't in the dataset.
Any related open or closed issues to this feature request?
#9002