Introduce Benchmark utility to support running different SDK versions

[maschad]

If I would like to measure performance improvements/regressions between two TS SDK versions, we currently have no means of doing that. Ideally we could have a tool that would spin up an instance of two versions of the SDK and perform benchmarking tests such as sending a transaction or deploying a contract

[arboleya]

I think this could use a similar approach to how we do diffs of test coverage: by having two sets of results, one for the master branch and one for the current PR/commit—something like that. If we can upload/retrieve master bench results, we don't need to run things twice

[maschad]

I've done a bit more digging and we may not actually need such a tool as CodeSpeed can give us a historical overview, here's an example of sway

and here's an example of a historical overview of the compile step

If we are satisfied with this I think we can close this issue. Thoughts? @FuelLabs/sdk-ts

[Torres-ssf]

@maschad Nice, this seems to solve the problem.

No opposition from my side to close this issue.

[arboleya]

Agreed.