Testing suite for builder

Rationale

To allow anyone to easily make changes to the builder, measure their impact, and ensure that they are safe.

Implementation

Provide standardized tools and data (sample bundles)for testing
Provide a framework for testing (e.g. performance metrics)
Create guide to deploying and testing the builder

One way to approach this is to create a mass conflicting bundles at random and test merging against those. But thinking it through we might want a standardized dataset that allows for more precise measurement of performance (e.g. variance in the randomized generation of bundles doesn't affect performance)

I was thinking we need these things:

provide a guide to setting up the builder for local development
provide a test script to run
provide testing data (bundles and txs)

the output should be something like:

hey, the code all works as intended and these X number of checks were met!
XX bundles were merged
XX bundles were simulated
in XX time XX block templates were generated
the most profitable block was XY ETH with XX bundles merged alongside XX transactions

ideally this is somehow standardized with the same inputs so that you can directly compare 2 versions of the code and say "hey, this version performed better than the other, which performs better than the main branch"