Investigate quinn performance on main and parallel-transmit branches

[flub]

As we are starting to move to working on quinn we need to check out the baseline of quinn's main branch. This involves porting iroh to those branches and then we can get netsim numbers.

Early tests indicate main might have a regression for 1-N transmission, but an improvement for 1-1. The parallel-transmit branch is supposed to improve on that. So we need to figure out if that change is enough or if we still need to understand where the 1-N perf regression came from on main.