Generates Localize.Number.Lexer.Categories, the Unicode general
categories Localize.Number.Lexer classes characters by, from the
unicode library.
The lexer settles most characters with its lexicon: digits, signs
and separators. What is left it classes by Unicode's general
category: a format character (Cf), a digit of another number
system (Nd), a punctuation mark (Po or Pf) or a space (Zs).
Those categories change with each Unicode release, so the table is
generated from the unicode library, which tracks them, and checked
in. Run this task when that library updates:
mix localize.gen_unicode_categoriesThe unicode dependency is a development one; the generated module
needs nothing at runtime.