Set up cluster benchmarks to automatically validate benchmarks

[misiugodfrey]

Benchmark validation is supported by our scripts, but requires an explicit path to be set for the results set.

The scope of this item is to point at a expected data directory where the expected results should be detectable by a schema name similar to how we detect data sets. This is mainly to set up in the cluster scripts.