This is very much incomplete and not fully tested. Likely still a
long way to go. At least it works with a some simple stuff. However,
tails are currenlty processed twice and various now-uneeded pieces
have not yet been removed.

Added a new treeprocessor which runs inlinepatterns without the need to use
placeholders or alter patterns. Note that any pattern that made use of
groups will need to be updated (reduce each group by 1) as a result.

The smart_strong extension has been removed and its behavior is now the
default (smart em and smart strong are the default). The legacy_em
extension restores legacy behavior.

This completes the removal of keywords.  All parser behavior is now modified
by extensions, not by keywords on the Markdown class.

If you have existing documents that use the legacy attributes format,
then you should enable the legacy_attrs extension for those documents.
Everyone is encouraged to use the attr_list extension going forward.

We started with the numbers before HTML5 was a thing and we thought there
might be an XHTML2. Today, we know that all we have are HTML style tags and
XHTML style tags. Nothing else really matters in the real world.

Note that if '(x)html1' '(x)html4' or '(x)html5' are passed in, the number is
stripped/ignored. Users shouldn't need to change their code for this.