frontend, libobs: Fix volume control state desync

[Warchamp7]

Description

This PR accomplishes two goals:

1. Fixes the volume control button state getting desynced when using plugins/scripts to change mute/monitoring
2. Makes it so monitoring while muted is handled directly in libobs instead of the frontend

Motivation and Context

1. This fixes Audio Mixer UI buttons desync when controlled via WebSocket #13246. libobs sends the signals for the relevant state changes before the state is actually applied to the object. This means if you respond to a signal for mute by calling obs_source_muted() you are going to get a stale value at best or a race condition at worst. Instead the VolumeControl now updates a local value using the new value from the signal, so that the up to date value can be used for updating the mixer state.

2. Currently in the new audio mixer, the ability to hear a monitored source when it was muted was handled via frontend logic. This logic would consider the UI input and only actually mute a source if monitoring was disabled. This led to some inconsistencies when things like the Advanced Audio Properties window or websockets tried to update these settings, as they would not flow through that logic.

This approach was taken as I naively assumed that a muted source would never even make it to the part of the audio pipeline where monitoring would be considered. However after further investigation thanks to @PatTheMav, we gleaned that monitoring itself is done via audio callback and had specific handling for the muted state already.

How Has This Been Tested?

Changed the mute and monitoring state of multiple audio sources via the mixer, the Advanced Audio Properties window, and via obs-websockets.

Types of changes

• Bug fix (non-breaking change which fixes an issue)

• Code cleanup (non-breaking change which makes code smaller or more readable)

Checklist:

• I have read the contributing document.
• My code has been run through clang-format.
• My code follows the project's style guidelines
• My code is not on the master branch.
• My code has been tested.
• All commit messages are properly formatted and commits squashed where appropriate.
• I have included updates to all appropriate documentation.

[PatTheMav]

I'm surprised this works, because that's usually a C++ feature, so maybe a GNU extension?

[RytoEX]

We do have C extensions enabled for Linux because libobs doesn't compile without them there.

obs-studio/cmake/common/osconfig.cmake

Lines 5 to 17 in 74efa22

	if(CMAKE_HOST_SYSTEM_NAME STREQUAL "Windows")
	set(CMAKE_C_EXTENSIONS FALSE)
	set(CMAKE_CXX_EXTENSIONS FALSE)
	list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/windows")
	set(OS_WINDOWS TRUE)
	elseif(CMAKE_HOST_SYSTEM_NAME STREQUAL "Darwin")
	set(CMAKE_C_EXTENSIONS FALSE)
	set(CMAKE_CXX_EXTENSIONS FALSE)
	list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/macos")
	set(OS_MACOS TRUE)
	elseif(CMAKE_HOST_SYSTEM_NAME MATCHES "Linux|FreeBSD|OpenBSD")
	set(CMAKE_CXX_EXTENSIONS FALSE)
	list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/linux")

[PatTheMav]

Yep we need that for stuff like POSIX threads to work IIRC, was just wondering if that's the reason why that works on Linux, because it won't on macOS and I'm not sure whether MSVC accepts it either.