We always have a swapchain here, even if there are no back buffers.

Depending on the location flags, the buffer or heap_memory should be
used.

This check is even more inconsistent between versions. Notice how the v2
interface is particularly broken because it never AddRef()'s the new iface.
This check also seems to be the original source for the questionable behaviour
in d3d_device_set_render_target() of storing the new iface before checking the
result of the wined3d_device_set_render_target(). (In particular,
SetRenderTargetTest() in tests/d3d.c tests for this behavious.)

This is where things become a bit of a mess, because the error checking is
inconsistent between versions.

winemac: Use NSMouseInRect() instead of NSPointInRect() to compensate for Cocoa's off-by-one coordinate system.

This matches what Cocoa does when determining how to handle an event so that,
for example, our test if a click is in the window grow box corresponds to
whether Cocoa will run an internal mouse-tracking loop to resize the window
when we pass it the event.  This fixes a problem where both Cocoa and user32
would try to run a resize loop and the cursor would get "stuck" resizing the
window after the button was released.