What's the main difference / addition to the databricks' spark-sql-perf?

[clehene]

I'm asking because I noticed it leverages that, but at the same time copies some things over - is there any benefit from using this?

On a different note it may be worth clarifying the differences / additions in the README