Client-side ingestion

[davenquinn]

We have a pretty nice task-based ingestion process now. However, it is tedious to run these on the server in an unfamiliar Python environment. We should create a new method for ingesting data into Sparrow with provisions for either working with files already known to Sparrow or uploading new files through the CLI.