This marks the beginning of the implementation for Codec's AddmusicK Beta. Note
that the upcoming code will be sourced, though also probably slightly modified,
from the C++ source, which is otherwise not public, unlike the rest of the
extant beta: I acquired the source from Codec themselves.

This branch will not be merged until all of the music changes are implemented.

This commit mentions #216.

Mostly adapted from the source for Codec's AMK beta, though with some minor
adjustments.

This commit mentions #216.

Mostly adapted from the source for Codec's AMK beta, though with some minor
adjustments.

This commit mentions #216.

Mostly adapted from the source for Codec's AMK beta, though with some minor
adjustments.

This commit mentions #216.

Pretty much directly adapted from the source for Codec's AMK beta, though with
some minor adjustments.

Resetting key on and non-key on remote code events are not yet implemented.
Technically they can be since I have them in another branch, but that still
requires some inheritance, so to speak.

This commit mentions #216.

Some of these require new hex commands or SPC700-side code. Others, well... they
simply haven't been implemented yet.

My bad for this failed circumnavigation (and not thinking things through with
using make on this before pushing).

This one stops the parser from guessing the tempo, as it already does not guess
when any tempo fade is used.

An erroneous text position increment was added on where it wasn't supposed to be
since it was already done, causing one of the digits to not get read correctly.

…o check

The source code I got made a mistake with testing with ltempo, when ltempo was
not the variable that was supposed to be checked. Also, AddmusicK used a tabbed
block without using brackets, which for me is more Python-esque than C++-esque.

A bad variable reference was used, and the v commands were outputting the wrong
values.

With the addition of the two remote code events tailored for the two types of
events, the implementation of (!!) can be completed on a technicality.

…AMK Beta

(!!) was initially not going to be implemented in parser version 4 due to me not
wanting to grab anything that required a new hex command. However, due to the
way AMK Beta's Remote Gain was restored, this MML command can now make the cut
into parser version 3.

: and % are still cut from parser version 4, on that note: : because it requires
a new hex command, and % because of a conflict with binary notation in asar (I
have decided to reserve this for parsing a binary number).

Partially cherry picked off of commit e026687 .
Does not include components that currently require a different commit to
implement.

Didn't realize a marker was removed in the process of merging, which would have
caused a compilation failure.