When running a migration package where only some of the files fail to be migrated for some reason, it would be really useful if I could rerun the same migration only processing those items that failed in a previous run.