feat(benchmark-tool): Proof of concept for custom benchmarks

[alexvy86]

Description

This is a proof of concept adding a feature to the benchmark-tool, so we can use it to write benchmarks that report their own custom measurements.

Reviewer Guidance

The review process is outlined on this wiki page.

Now with a third kind of benchmark (execution time, memory usage, custom outputs) some patterns where we could be reusing code across all kinds become clear but I didn't try to refactor in that direction for the proof of concept.

Interested in feedback about the new feature, if the API seems good enough, where we might need more flexibility, etc.

[Abe27342]

API seems fine to me. I also think it's not really a big deal if we decide we want to tweak it a bit in the future, since we don't have specific cadences for updates to this package.

[microsoft-github-policy-service[bot]]

This PR has been automatically marked as stale because it has had no activity for 60 days. It will be closed if no further activity occurs within 8 days of this comment. Thank you for your contributions to Fluid Framework!