This reduces pdf file sizes when usetex is active, at the cost ofsome complexity in the code. We implement a charstring bytecodeinterpreter to keep track of subroutine calls in font programs.Give dviread.DviFont a fake filename attribute and a get_fontmapmethod for character tracking.In backend_pdf.py, refactor _get_subsetted_psname so it calls a method_get_subset_prefix, and reuse that to create tags for Type-1 fonts.Mark the methods static since they don't use anything from the instance.Recommend merging to main to give people time to test this, not toa 3.10 point release.Closes#127.Co-Authored-By: Elliott Sales de Andrade <quantum.analyst@gmail.com>

Co-authored-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>

Co-authored-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>

The old optimization where the same font file and descriptor could beshared by multiple differently-encoded fonts is wrong in the presence ofsubsetting, unless we also take the exact glyph subset into account.It is very unlikely for the exact same subset to be used by differentencodings, so just remove the optimization.Rearrange _embedTeXFont in neater blocks.

The DejaVu and Heuristica fonts are used by the type-1 fontsubsetting tests.Heuristica has a Cyrillic encoding and apparently cannot beloaded without installing texlive-lang-cyrillic.

Type-1 font subsetting