Go sometimes subsitutes literal character codes (e.g. '\u2206') with other runes. This doesn't
happen if the literal is written as a number (e.g. 0x2206 ). I am not sure why Go does this.
These maps had been getting messy which made it hard for me to update them.
This change builds the full maps from source files which are included in the commit.
Made GlyphToCode work for all tables
Moved more aliases into glyphAliases rather than leaving the duplicates in the base maps.
Use SimpleEncoder explictly for simple fonts
- Added Type1 font parsing.
- Added Standard 14 font parsing.
- Fixed some bugs in cmap code.
- Started re-structuring of font code. Moved common font fields to `fontSkeleton`
