Update: measure plugin loading time and output in debug message

[victor-homyakov]

What is the purpose of this pull request? (put an "X" next to item)

[ ] Documentation update
[ ] Bug fix (template)
[ ] New rule (template)
[ ] Changes an existing rule (template)
[ ] Add autofixing to a rule
[ ] Add a CLI option
[ ] Add something to the core
[X] Other, please explain:

What changes did you make? (Give an overview)
I've added measurement of loading time for each plugin. Times are shown in debug output.

The reason is in unexpectedly significant startup time for some plugins, even when there is nothing to check for them. One possible example:

• ESLint configuration has typescript-eslint plugin added
• pre-commit hook is configured to run ESLint on modified files
• the commit does not contain .ts files (only .js, so ESLint is still launched, but there is nothing for typescript-eslint to check)

In this example typescript-eslint plugin is still loaded and consumes more than half a second to load typescript-eslint/typescript-eslint#1040

More examples:

• Performance: plugin adds 150-200ms to ESLint startup time because of ramda lo1tuma/eslint-plugin-mocha#213
• eslint-plugin-react - adds about 250ms to ESLint startup time
• eslint-plugin-import - adds about 350ms

So in case you added these four plugins, ESLint will require 1.5s to start. It's significant for both CLI tools and for pre-commit hooks. I'd like to know exact timings in order to tune ESLint configuration (pre-commit hook may have the fastest one, CI configuration should have the slowest one).

Is there anything you'd like reviewers to focus on?

[jsf-clabot]

All committers have signed the CLA.

[platinumazure]

LGTM, thanks!

We usually ask for an RFC before most core changes, but I think this is trivial enough that we shouldn't need one here. But let's see what other team members think.

[platinumazure]

I should qualify: This looks good, outside of the CLA check. Please let us know if you run into issues while fixing the commits to allow the CLA check to pass. Thanks!

[ilyavolodin]

LGTM. I don't think we need RFC for this.

[platinumazure]

Marking as "evaluating" for now, but we can change to "accept" if nobody from the team objects to this change.

[g-plane]

LGTM, thanks!

[platinumazure]

I'm marking this as "accepted", since the team has had plenty of time to review and there is no objection so far. I'll leave this open for another day or so, and then merge.