Attempt to regenerate a ToUnicode map using embedded font data
as a key.
Complex fonts are generally embedded using an identity encoding
which references characters by glyph index rather than character
code. Because glyph indexes are specific to a particular font, to
extract text from the PDF you need a table mapping the glyph
indexed to character codes. This map is called the ToUnicode
Well behaved PDF producers will insert a ToUnicode map for all
fonts that they embed. However some font producers may fail to do
so. This means that copying or extracting text from such a PDF will
result in gibberish.
In some cases it is possible to regenerate the ToUnicode map
from the embedded font data. This function attempts to do so and
returns true if a new ToUnicode map was inserted. This function
only works on composite fonts and calling
it on other types of fonts will not do anything.