[core] Align prettier tooling with main repository

[oliviertassinari]

The only difference left is the print width, as detailed in #6, which would be better to handle in a follow-up pull request, to make this one easier to review.

Regarding Lodash, we might want to ignore the prettier update coming from it, until we remove it from the source: #7.

[dtassone]

Prettier currently run with lint that's why there is no individual script
you do a yarn lint it applies linting and prettier

Also we exclude /lib as it contains vendor libraries so lint there

[oliviertassinari]

you do a yarn lint it applies linting and prettier

I wasn't aware that this was possible. It seems to be documented in https://prettier.io/docs/en/integrating-with-linters.html#use-eslint-to-run-prettier. Cool, I have learned something.

Going back to my previous point on sync all the tooling between https://github.com/mui-org/material-ui, https://github.com/mui-org/material-ui-x, and https://github.com/mui-org/material-ui-pickers. In our case, I think that core's approach has the benefit of:

• Formating markdown and JSON formats (you can see the JSON diffs in this pull request).
• Only formating the files that changed since your last iteration (compared to the baseline git branch). As the codebase will grow, eslint will be too slow to run after each pull request in the local env, to format it before running the CI. Try doing that on the core repository to experience the issue (takes 1m+ time yarn lint).

[oliviertassinari]

Ok, this time, we should be good. I have manually listed the lib folders so it gives us an extra incentive to progressively remove them.

[oliviertassinari]

@dtassone I have updated prettier to lint the JSON and Markdown at the same time.