snprintf() is available on mingw, and the libnfs kludge broke the
build with mingw, because sprintf_s() was now both an inline function
and a "dllimport" function (because the macro renamed the inline
function snprintf() to sprintf_s() in mingw's stdio.h).

Fixes regression by commit 23d5a2b8 -
that commit always pretended that any Opus file has both track and
album gain, and thus disabled the fallback to the other if one is not
set.

This patch changes the logic to only submit ReplayGain if at least one
value is set, and apply the offset only to that value.  If none is
available, then the new check in HandleAudio() will submit only the
output gain.

Closes https://github.com/MusicPlayerDaemon/MPD/issues/977

By bit-wise ANDing the reported flags with GetScheduledFlags(),
ERROR/HANGUP always get cleared.  This means the MPD event loop could
never report those conditions.

This just happened to break the Windows build because of the
`GetObject` macro in `windows.h`, so I added a kludge to
PollResultGeneric.hxx.

Oh no, 3413d1bf was broken!  Instead of passing a number as
"seconds" to the duration constructor, it just abused the duration
constructor as cast operator, which caused custom state_file_interval
settings to be extremely short.

Don't attempt to initialize the io_uring subsystem more than once.

On Windows, there is an IsWhitespaceOrNull() overload with TCHAR, and
the compiler doesn't know which one to pass to std::find_if().

Don't copy the input StringView.

Closes https://github.com/MusicPlayerDaemon/MPD/issues/974

cue_next_word() can return a pointer one past the end of the string if
the word is followed by the terminating null byte.

This commit adds some basic infrastructure for fuzzers, and adds a
fuzzer for the CUE sheet parser.