GRT improvements

[martenson]

first take on stats to gather:

• histories (count, avg size, %shared, %published)
• workflows (count, invocations, assess complexity, imported?)
• jobs (states, counts, metrics)
• users (count, local x remote)
• pages counts
• library counts
• visualization counts
• dataset (count, types, size)
• write this into a separate GRT script / "safe" data to send.
• send data to an influxdb so we can make dashboards @erasche