Complete Ingestion of Docker Images

[10d9e]

Create jobs to continually ingest images from target docker registries.
Given the volume of images, the ingestor should have the following:

• automatic error recovery/retry logic
• persistent repository upload tracker - badgerdb or similar db to track the progress of uploads
• explore "batched" upload outside of registry (may not be possible, but worth looking at)

[10d9e]

This process is quite slow through the native docker client. There may be a better way to rewrite large-scale ingestion