Unicode support in identifiers

Just debugged my way to this comment:

// Note: In addition to letters, numbers, underscores, and dollar signs, modern ECMAScript
// also allows Unicode categories such as letters, combining marks, digits, and connector punctuation.
// These are mostly supported in all environments except IE11, so if someone wants it, we would accept
// a PR to allow them (although the test surface might be somewhat large).
StringChecks._identifierBadCharRegExp = /[^a-z0-9_$]/i;

I mark @internal identifiers with a trailing Δ so I'm hitting a few issues with this! I'm happy to attempt a PR, but I'm curious what prior work has been done towards this, and what would be expected in terms of test coverage. I believe for my usecase there may have to be changes within the Tokenizer as well ({@link} parsing is a bit broken too), seems like there's plenty room here for stepping on toes 😅

Let me know what you reckon?

Start with StringChecks._identifierBadCharRegExp and the Tokenizer, especially the reported {@link} parsing path. Review prior work and define the required Unicode identifier categories and test coverage before changing behavior. Done means supported Unicode identifiers parse correctly without breaking existing identifier or link parsing.