Proposal: Simplify spec's whitespace handling

[sanbox-irl]

The yarn spinner spec currently defines whitespace as follows:

Whitespace is any non-visible character with a width greater than 0. Common whitespace encountered include the space and the tab.

This allows for valid, meaningful non-zero width characters, especially bare accent markers, to be considered "white space". The arabic letter mark 061C, for example, is whitespace according to this definition but not according to anyone else.

Instead, since Yarn already requires its input to be valid utf8, I suggest we punt to UTF8's own definitions of whitespace, found here: https://www.unicode.org/Public/UCD/latest/ucd/PropList.txt.

This is precisely what C#'s (https://learn.microsoft.com/en-us/dotnet/api/system.char.iswhitespace?view=net-7.0#remarks) white space already does, so we're in good company.

Therefore, I would suggest that the line be chnaged to:

Whitespace is, generally, any non-visible character with a width greater than 0. Common whitespace encountered include the space and the tab. Specifically, whitespace is specified in the [Unicode Character Database](https://www.unicode.org/reports/tr44/) [PropList.txt](https://www.unicode.org/Public/UCD/latest/ucd/PropList.txt).

[McJones]

I am totally in favour of this, the Unicode consortium has already done the hard work and we should take advantage of this.

[sanbox-irl]

I'll make a PR tonight then