Automatically run go benchmarks

[mhutchinson]

We have a few benchmarks for critical code in this repo. These benchmarks aren't run in CI. An ideal workflow would allow PRs to be flagged if they cause regressions in these benchmarks. We would also be able to see how the benchmarks have changed over time.

https://github.com/benchmark-action/github-action-benchmark promises to do this.

[mhutchinson]

Cool, looks like we have some benchmark data now:

https://transparency-dev.github.io/trillian-tessera/dev/bench/

This meant setting up a permanent branch in this repo at gh-pages to host this data.