`tests_alternate`'s Arabic tests look up installed fonts by name
(`FontDb::face`), which loads each one into the global database and so
changes its generation, without `tests_fontchange::adding_fonts()`.
When that lands in the first frame of
`tests_charstroke::character_stroke_outlines_are_kept_between_frames`,
the second frame lays the type out again and expands the stroke again,
and the test fails (3 runs in 30 on macOS, where Arial and Geeza Pro
are installed; where no Arabic font is, those tests return early and
nothing changes). They now hold the lock, and its documentation says
looking up an installed font counts as changing the fonts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Highlight Alternate Glyphs (8689863f) flagged any glyph that isn't its
character's cmap glyph, so the joining forms Arabic and other complex
scripts always take (init, medi, fina, isol, rlig, ccmp and the Indic
required set) covered ordinary text in pink. A glyph is now an
alternate only when required shaping alone doesn't give it: the text is
laid out again with every optional feature off (liga, clig, calt and
the features its characters turn on) and each candidate glyph compared
by cluster. The pink boxes are measured once per text geometry, only
when the preference shows them.
