[docs] - Add benchmark methodology section to Readme

[astuyve]

Description of changes: When I first received access to this project I was very pleased to see that the benchmarks measured round-trip request latency for a given cold start.

I don't believe this is common knowledge (as it surprised a number of folks when I spoke about it at re:Invent), so I believe a short description is helpful.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

I agree!