Spike: copy S3 database/schema data to another location in S3

[brianhw]

Support copying Mysql database data on S3 from one location to another. Do the same for Vertica schema. Main purpose is to move S3 data to be used as a "reference" to a separate location in S3 from that used for daily production runs.

This is mostly for demonstration at this point, since it requires installing awscli on the cluster.