Support unicode character categories

Regex allows [unicode character escapes](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Regular_expressions/Unicode_character_class_escape), however axe-core currently cannot use them as we still support IE11. This was noted in our `unicode.js` file where we [wanted to use the CF category](https://github.com/dequelabs/axe-core/blob/develop/lib/commons/text/unicode.js#L88-L92).

Supporting the [new label-content-name-mismatch algorithm](https://www.w3.org/WAI/standards-guidelines/act/rules/2ee8b8/proposed/#label-in-name-algorithm) calls for using the characters L and N to filter out everything else in the string. It will be much easier and more sustainable to write those using regex character escapes, but for now we have to hand write the same thing in unicode characters.

Start in lib/commons/text/unicode.js, especially the CF-category note around lines 88-92, and read the linked label-content-name-mismatch algorithm to understand the L and N filtering need. Check how axe-core currently represents these Unicode ranges and how IE11 compatibility constrains the approach; done means the required character categories can be used sustainably without dropping IE11 support.

Written by the indexing model from the issue text.