chore: reorganize benchmarks

[JLHwung]

Q	A
License	MIT

The benchmark files scattered in babel-generator, babel-helper-validator-identifier and babel-parser are now moved to a top level /benchmark folder. They now share the same /util.mjs benchmark setup and we don't have to hardcode package index.

[JLHwung]

Now we can compare performances between any versions of babel packages by revising this package.json.

[KFlash]

what's the perf diff between Rome and ESbuild vs Babel? Equal in performance?

[codesandbox-ci]

This pull request is automatically built and testable in CodeSandbox.

To see build info of the built libraries, click here or the icon next to each commit SHA.

Latest deployment of this branch, based on commit a6275b2:

Sandbox	Source
babel-repl-custom-plugin	Configuration
babel-plugin-multi-config	Configuration

[babel-bot]

Build successful! You can test your changes in the REPL here: https://babeljs.io/repl/build/47585/

[fedeci]

Do we have any report tool that runs benchmarks along with tests on the CI?

[JLHwung]

Do we have any report tool that runs benchmarks along with tests on the CI?

Not yet. After we centralize the benchmark files, we can come up with a benchmark bot that plots benchmark alongside versions so we can catch performance regressions from new releases.