Maybe the `HIGH_WATER_MARK` idea doesn't make sense?!

Maybe we should simplify the code, so we just fill the io_uring submission queue until `len(sq) + len(cq) - MAX_ENTRIES_PER_ITERATION >= SQ_LEN`. Don't bother with the `HIGH_WATER_MARK` idea. I'm not sure it makes sense?

I should probably benchmark though!