decklink-ui: fix double free of settings

This leads to a crash when the output is stopped, either by clicking the
stop button or when exiting OBS studio.

This crash is:
- intermittent in release builds
- reliable for me in debug builds

It is held by an `OBSData`, which, if not null, will automatically
call `obs_data_release()` on scope exit.

obs-transitions: Track Matte support for Stinger Transitions

This adds the ability to use a secondary black-and-white video as a mask
between source A and B of the transition. The greyscale value of each
pixel is used as the "slider" value in a linear interpolation between the
corresponding pixels in source A and source B.
The track matte can either be in the same file as the stinger itself
(next to the stinger or under the stinger, doubling the width or height
of the stinger depending of the selected layout) or a in a separate
dedicated file.
The same file/separate file behavior is controlled by the
"Matte Layout" option in the stinger settings.

Fixes a bug where the notify object existed after source deletion

Probably long time coming, but when the user changes their default audio
device in Windows sound settings, OBS will now detect it and change the
audio device automatically to the new device if it was set to use the
"Default" device.

Use _ARM_BARRIER_ISH for ARM path.

Remove useless conversion in os_atomic_load_bool.

Use native instructions on ARM64 (dmb is slow).

Use magic number for 32-bit ARM since _ARM64_BARRIER_ISH is for ARM64.

Makes the crashlog a little nicer to read.

Use function that updates previous value automatically.

Also load initial value seq_cst to be safe.

Add exchange functions to alias the poorly named set functions.

Add store without reading previous. Faster on non-x86 processors.

Add compare-exchange that updates previous to avoid redundant fetch.

On Windows, load bool without conversion from char.

On Windows, load using mov with compiler barrier. Still seq_cst.

On POSIX, use GCC __atomic builtins.

df4eb821 fixed a bug that caused source audio timestamps to perpetually
lag. However, there is a deeper issue where after we reach max
buffering, lagging sources make OBS's entire audio pipeline fall over.
These may be corrected by later code, but still cause global audio
glitches at best. Persistent problems, as prior to df4eb821, cause audio
to fail entirely.

The root cause is that OBS's audio mixing tree cannot deal with
timestamps prior to the current audio tick. Intermediate mixing stages
assume that the lowest incoming timestamp is the base of the current
tick, and mix accordingly. This propagates lagged timestamps up the
tree, where at the top level mix_audio will drop the source entirely -
which at this point is a transition covering all inputs, thus glitching
audio globally. Where extra buffering can cover the slip, the entire mix
gets retried and the error corrected, but when the global buffer
duration is maxed out, it makes it to the output.

The solution is to catch laggy sources immediately after rendering, and
drop audio to bring them back in sync, or mark them pending if not
enough audio is available. This ensures later mixing stages are not fed
with out of sync timestamps.

This improves the ignore_audio code to only drop as much audio as
needed to bring the source back in sync, and moves its call to
immediately after source audio rendering.

This is a safety against sources which somehow end up with
audio_pending=false but ts=0. Other codepaths guard against this too.

This adds additional capture presets, including 3840x2160 and
1920x1080, in addition to the preset "High." These are guarded with
a runtime check using the @Available() keyword for macOS 10.15+.

linux-capture: Fix Problems with the Window-Selection of the XComposite Source

Bug is caused by the internal connection variables not being reset on
reconnect, leading OBS to both be unable to parse valid packets from and
send valid packets to the remote end.  This commit splits RTMP_Init off
into a new RTMP_Reset function, which resets these internal variables
without re-initing the rest of the library.  The original RTMP_Init
calls the new function, perfectly preserving the old behaviour while
adding a new reset function to address the issue with.

Fixes obsproject/obs-studio#2865

libobs + UI: Add Missing Files API & Dialog

Add display capture via Windows Graphics Capture

Another warning clean-up pass

While the current code only ever calls try_connect() with the input
argument 'device' in the range of 0 and MAX_DEVICES, this adds a check
to ensure that future code does not break the following sprintf.

In addition, use snprintf instead of sprintf to ensure that if anything
breaks, the sprintf does not lead to memory corruption. Again, the new
check should already make sure of that, but the additional effort of
using snprintf instead of sprintf is so low that it is worth to have a
little more security in the future.

Removes prior attempt to expose libcaption headers which really shouldnt
have public. This instead moves the obs-internal include out of the
public obs-scene.h and into it's implementation.