polish(ast): prefer undefined over empty arrays

[yaacovCR]

see: #2405 (comment)

Manually created ASTs always allowed undefined in place of empty arrays; this change simply updates the parser to more closely follow the manual approach. This is therefore not technically a breaking change, but considering that there may be tools not aware of this, we have labelled it a BREAKING_CHANGE to highlight it for consumers.

This closes #2203 as our tests now cover the undefined case by default whenever there is an empty array.

[netlify]

✅ Deploy Preview for compassionate-pike-271cb3 ready!

Name	Link
🔨 Latest commit	125c2b7
🔍 Latest deploy log	https://app.netlify.com/sites/compassionate-pike-271cb3/deploys/670cf48b36880d0008726e05
😎 Deploy Preview	https://deploy-preview-4206--compassionate-pike-271cb3.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[github-actions]

Hi @yaacovCR, I'm @github-actions bot happy to help you with this PR 👋

Supported commands

Please post this commands in separate comments and only one per comment:

• @github-actions run-benchmark - Run benchmark comparing base and merge commits for this PR
• @github-actions publish-pr-on-npm - Build package from this PR and publish it on NPM

[yaacovCR]

Well, it works!

It would seem that it's more accurate to use undefined when an optionalMany(...) is skipped entirely and an empty array when it is found, but has an empty list. So, it would seem that this PR is technically more correct.

Peeking around at https://astexplorer.net/ it seems that TS parser has some empty arrays, some undefined, while for Babel I didn't on a very cursory look see any undefined.

But still not 100% sure this is too compelling, with the status quo being somewhat compelling! Comments welcome!

[yaacovCR]

Merging this PR allows us to close #2203, improving robustness of our test coverage. Presumably our downstream users were also affected by #2203 and had to account for manually created AST nodes, so they should be robust to this change, but I upgraded it to a breaking change anyway just to call it out.

[JoviDeCroock]

Probably needs a rebase