update to Unicode 18.0.0

[anthrotype]

This updates the generated tables to Unicode 18.0.0 and backports CPython's derived-name ranges (python/cpython#144789, python/cpython#157634), adding name()/lookup() support for Tangut, Jurchen and Small Seal characters.

[anthrotype]

Most of this diff is regenerated Unicode tables. The version bumps, checksum update and adjustments to existing test code points are mechanical too.

The part worth reviewing more closely is the derived-name handling in makeunicodedata.py and unicodedata.c, backported from python/cpython#144789 and python/cpython#157634. This replaces the hard-coded CJK ranges with generated ranges and extends name()/lookup() to Jurchen and Small Seal, also fixing the older Tangut naming gap. Just regenerating the tables with the old code would leave these names missing.

The local adaptations use UNICODEDATA2_TOUPPER and a small ASCII prefix-comparison helper for PyPy compatibility. I also ported additional character-name tests from CPython, including its comparison against DerivedName.txt (downloaded from Unicode.org during test setup).

[anthrotype]

the older Tangut naming gap

Technically this isn't specific to Unicode 18: with Unicode 17, name('\U00017000') raised ValueError and lookup('TANGUT IDEOGRAPH-17000') raised KeyError. But backporting CPython's derived-name handling fixed both.