Modeled after std::filesystem::operator/() from C++17.

This is just the API design; there is no implementation yet, and no
caller.

This method was added in Boost 1.58.

Not updating m4/ax_boost_base.m4 because the new version breaks
cross-compilation by adding -I/usr/include to BOOST_CPPFLAGS.

Closes #296

Some Audio plugin (such as ALSA, and soon CoreAudio) already support microsecond level buffer time. However, current interval less than 1000 microseconds will cause a bug that the code treated as 0 ms.

Moving some text from the manpage.  Oh, documentation is still a mess.

Prepare for per-listener permissions.

some device seems to have issue with setting kAudioDevicePropertyVolumeScalar  with kAudioObjectPropertyElementMaster. Use AudioToolbox 's kAudioHardwareServiceDeviceProperty_VirtualMasterVolume instead. 

Ideally, we should get the steoro channels first, and set the kAudioDevicePropertyVolumeScalar for each channel, which is doable as presented in https://github.com/cmus/cmus/blob/master/op/coreaudio.c. I will do a follow up PR after refactor PR.

1 sec for pause is too long. we wait for the same amount of time as when ring buffer is not available for writing.

Remove sync_sample_rate option as it's no longer necessary

This PR will fix #271.

special thanks to @coroner21 who contributed a nice way to score hardware supported format in #292

Also, The DSD related code are all guarded with ENABLE_DSD  flag.