Certain devices (particularly certain mixers and soundflower 64ch) would
have an arbitrary number of channels, and wouldn't really be mappable to
a specific speaker layout supported by libobs.

So to fix this issue, if the channel count is above 8, force the data to
stereo to ensure playback can still occur, rather than cause it to just
fail.

Prevents issues on specific locales, especially where decimal points are
represented by commas rather than periods.

This is just in case a filter is applied to a scene, filters will turn
off the blending so it's necessary to set the blending.

This optionally specifies that the int/float value should be displayed
with a slider rather than and up/down control.  UIs are not necessarily
required to implement it, it's meant to be more of a hint.

This fixes an issue primarily with filter rendering: when capturing
windows and displays, their alpha channel is almost always 0, causing
the image to be completely invisible unintentionally.  The original fix
for this for many sources was just to turn off the blending, which would
be fine if you're not rendering any filters, but filters will render to
render targets first, and that lack of alpha will end up carrying over
in to the final image.

This doesn't apply to any mac captures because mac actually seems to set
the alpha channel to 1.

This returns a common effect useful for rendering an image with the
alpha channel overridden to 1.0.

If negative was specified it would not parse the negative sign and thus
would not interpret that number as negative, and would cause
shader/effect parsing to simply fail on the file.

This is particularly important for the filter pipeline in order to
ensure that when the last filter is reached that the original blend
state is properly reset.

The linux window and monitor captures draw in a custom way, therefore
they must be marked as types of sources that do custom rendering.

Switching between shaders often can cause a performance hit, so cache
effects so that they persist until the graphics subsystem is destroyed.

If a render target was switched from one to another and then back
consecutively, the texture would not get reattached at the last point
due to the fact that the cur_render_target variable of the FBO storage
structure would already be set to that texture, and then it would never
get reattached because it thought it was already using that render
target.

So to fix this, any time a render target legitimately changes from one
to another, it sets the cur_render_target and cur_zstencil_buffer
variables of the FBO structure to null.

I encountered a scenario where effects might reference shader parameters
that may not potentially exist, this isn't particularly a bad thing that
needs to return failure.  It just needs to gracefully ignore it.

When render targets are used, they output to the render target inverted
due to the way that opengl works.  This fixes that issue by inverting
the projection matrix so that it renders the image upside down and
inverting the front face from counterclockwise to clockwise.

Texture units were not being properly associated with the shader uniform
variables.

obs_leave_graphics() may not be properly called if this returns null for
whatever reason.

Mostly used for testing, partially green allows us to gauge whether or
not multiple filters are being properly applied

The wrong function was being used to recurse through the filter chain in
obs_source_process_filter, obs_source_get_[width/height] would get the
post-effect dimensions rather than the pre-effect dimensions.

The parameters order was wrong; sort of a rare thing to see, but it can
happen, especially when it's a function that's not used very often.

Certain callbacks and API functions were a bit vague on when and how
they were supposed to be used.

The frame is definitely meant to be modifiable if needed, so it
shouldn't be const.  I originally meant for these frames to be
duplicated, but with the source video cache that's both unnecessary and
would reduce performance.

Fixes a bug where get_format was overloaded by our own version
when forcing the codec to load with a HW decoder and not
reset to the original get_format if it failed to load.

Blackmagic DeckLink plugin

The code specific to Windows: helps convert `BSTR` instances to
`std::string`s; provides a Windows COM-specific implementation of
`CreateDeckLinkDiscoveryInstance`; aliases CFUUIDGetUUIDBytes,
CFUUIDBytes, and IUnknownUUID (the Linux SDK does this, but for some
reason the Windows SDK does not).

Some changes were made to the stock DeckLink SDK: removed all references
to legacy API headers in DeckLinkAPI.idl; removed all instances of
`importlib("stdole2.tlb");`.

The code specific to Linux helps convert `const char *` instances to
`std::string`s.

The code specific to Mac OS X helps convert `CFString` instances to
`std::string`s.

This is a cross-platform Blackmagic device capture plugin that makes use
of the manufacturer provided DeckLink SDK.

ComPtr can be used for things like the DeckLink API, but ComQIPtr has
windows-specific compiler features, so can't be used outside of windows.

Update hitbox.tv ingest servers list

In the same manner that PNG doesn't appear to work properly
with multiple threads, TIFF, JPEG2000 and WEBP also appears
to not render correctly (use of FFmpeg's ff_thread_* routines)
if decode is called before the automatic thread detection
has returned a suggested thread count for the decoder.

The reset method unnecessarily malloced a packet passed into
the packet buffer which does a dereferencing copy.

FFmpeg source updates: Clock chaining, frame dropping and friends

OBS_PROPERTIES_DEFER_UPDATE: Makes it so the properties are not updated
until editing of the properties is complete.