[subset] zero glyf and loca memory before writing so pads are clean. Test just the part of maxp we care about

[subset] Caclculate head table checksum adjustment

Test still fails, because we do not serialize tables in the same
order that fonttools subsetter does.

[subset] Fix several memory leaks in subsetting.

Also, call c++filt on their results.

Also build and fix and check harfbuzz-gobject.def

Also changes harfbuzz-gobject to distribute generated enum sources.

[subset] Add hdmx subsetting.

[subset] Fix checksum clearning in subset test diff. Add an updated expected file (has cmap4 restored).

More readable.

[subset] sketch out dropping in-glyf instructions. Sometimes yields differnet size glyphs than fonttools, possibly due to padding not being corrected