Automate benchmarking

[lightclient]

It would be great if there was a single command to benchmark the performance of the EE against several different scenarios (small tree, large tree, many transactions, few, etc) in Scout. This should also be run in the CI.

[lightclient]

#35 is just the first attempt at creating a strong suite of automated benchmarks for Sheth. I will keep this issue open until more work is put into it (specifically around testing the blob size + displaying the current results for the master branch on the README).