Signed-off-by: Connor McAdams <cmcadams@codeweavers.com>

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54960

Some applications create a IDirectInput and IDirectInputDevice instance
from their DllMain procedure, and starting the thread on the first user
creation and waiting for it to start will deadlock on the loader lock.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54916

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54917

Fixes: 8ae0c308
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54965

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54917

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54917

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54917

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54917

user32/tests: Do not stop checking the message list when skipping an optional or unsupported message.

This was either added by accident, or in an attempt to stop checking the message
list once we hit a winevent hook todo [and accidentally given too wide of a
scope]. However, the same commit also uses a global counter to ensure that only
one winevent todo is printed, so we don't need to break in the first hunk
anyway.

For trailing winevent todos, there is no global counter, so instead move the
goto inside the if block.

Fixes: a72bffe7

The test marked todo here was broken by 33617af8, which was hidden due to the above commit.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54955

And move its definition to ime.c.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54936

In d3d8, setting texture filters to invalid values does not yield an
error when calling IDirect3DDevice8::ValidateDevice. Some applications,
such as Worms Blast, rely on this behaviour.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54898

This test is an adaptation of the corresponding d3d9 test.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54898

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53947