Proposed: support 4-6 hex digits in character literal hex form

[DeLaGuardo]

The EDN specification currently restricts the hex form of character literals to exactly 4 hex digits (\uXXXX). This limitation prevents EDN from representing Unicode characters in the Supplementary Planes (U+10000 to U+10FFFF), which includes:

• Emoji (U+1F300–U+1F9FF)
• Historic scripts (Linear B, Egyptian hieroglyphs, cuneiform)
• Musical notation symbols
• Mathematical alphanumeric symbols
• CJK ideographs extensions
• And many other modern and historic characters

Extending the character literal hex form to support 4-6 hex digits would allow direct representation of any valid Unicode code point (U+0000 to U+10FFFF).

Many languages that can use EDN already support extended Unicode:

• JavaScript: Supports \u{XXXXX} syntax (ES6+)
• Python: Supports \UXXXXXXXX (8 hex digits)
• Ruby: Supports \u{XXXXX}

[mnemnion]

I'd like to see \u{...} adopted. Always found it very confusing that some languages mandate more hex digits than Unicode uses (six). It would mean there are two ways to write BMP codepoints, but: either that's true for the \U convention as well, or we'd have the egregious task of rejecting \U sequences which are "too small".

\u{...} doesn't add an additional escape character, it just adds another valid suffix rule to it. I would expect (hope!) that code would adopt the newer form over time for everything, and \uXXXX would persist as a legacy option. The \U convention makes encoders choose between wasting conditional branches figuring out which way to write an escaped point, or wasting bytes unconditionally writing it large. "u brace" does neither.