Add support of bulk redefinition of token renderers and token renderer prototypes in the \markdownSetup command using enumeration

[Witiko]

Continues #232 and #361.

Tasks

• Either parametrize \g_@@_glob_cache_prop with a sequence name or hardcode the sequence \g_@@_renderers_seq.

• Document that not just wildcards but also enumeration are supported and add an example for both.

• Make it possible to recover the renderer (prototype) name from the definition using #0.
  Here is how this can be done in expl3:

  \input expl3-generic
  \ExplSyntaxOn
  \tl_set:Nn \l_tmpa_tl { #0~#1! }
  \regex_replace_all:nnN { \cP\#0 } { Hello } \l_tmpa_tl
  \cs_generate_variant:Nn \cs_new:Nn { NV }
  \cs_new:NV \hello:n \l_tmpa_tl
  % Leave the text "Hello world!" in the input stream.
  \hello:n { world }
  \bye

  Document this possibility and add an example.

• Use enumeration and wildcards in tests (file keyval-setup.tex) in addition to the technical documentation.