Decoders might output some fps information and encoders might input fps,
but otherwise is unnecessary and may prevent compatible caps matching.

Better reflect its contents.

There is still a small amount of code in this file not directly related to the
FFP:

 - compile_state_table(), which should eventually go away;

 - states like state_so() and indexbuffer(), which should eventually be moved to
   context_gl.c with the rest of the resource loading/binding code.

The entire rest of state.c is concerned with the GL fixed-function pipeline.

device.c seems the most fitting place for these, since they're associated with
the device and stored in the device. device.c is also one of the few files that
contains client-side code, which these functions are.

Looks like touches are automatically released if the device doesn't send
updates regularly.

This should be more reliable than mouse moves and it exhibits more
issues with the rawinput buffer struct sizes on new WoW64.

Partial revert of 413fc34b.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=55421

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56244

It makes more sense to have it matching fx_5_0 format, which
uses the same flag for TBUFFER case, and also keeps 'single' modifier in there.
This modifier is ignored on fx_4_x, so it's only needed to be implemented
in the compiler.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56279

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>