Switch to the generic noncoherent direct mapping implementation.

This removes the direction-based optimizations in
sync_{single,sg}_for_{cpu,device} which were marked untestested and
do not match the usually very well tested {un,}map_{single,sg}
implementations.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMichal Simek <michal.simek@xilinx.com>