The change is needed because Asciidoc.py and Asciidoctor
do not process the '+' verbatim the same way. A span is
detected when the format sign (here '+')is preceded by a
non-word character. I haven't digged into the source,
but it seems that '{nbsp}' is considered a non-word sign
by Asciidoc.py, but not by Asciidoctor.
Using the double '+' opens 'unconstrained' span,
independent on the preceding character in both engines.
