If all completed requests in io_do_iopoll() were marked with
REQ_F_CQE_SKIP, we'll not only skip CQE posting but also
io_free_batch_list() leaking memory and resources.

Move @nr_events increment before REQ_F_CQE_SKIP check. We'll potentially
return the value greater than the real one, but iopolling will deal with
it and the userspace will re-iopoll if needed. In anyway, I don't think
there are many use cases for REQ_F_CQE_SKIP + IOPOLL.

Fixes: 83a13a41 ("io_uring: tweak iopoll CQE_SKIP event counting")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/5072fc8693fbfd595f89e5d4305bfcfd5d2f0a64.1650186611.git.asml.silence@gmail.comSigned-off-by: Jens Axboe <axboe@kernel.dk>