Avoid device to host copies by "partially declustering" certain nodes.

"Partial declustering" is defined as cloning a clustered node outside its
cluster and transferring some of its outgoing edges to the cloned version.

Some TensorFlow operations expect their inputs in host-memory and, because XLA
only produces device tensors, such nodes can incur a device-to-host copy if not
clustered along with their producers.  TensorFlow operations, on the other hand,
may produce their outputs in host memory so cloning the producer to outside the
cluster and moving the host-mem expecting consumers to use the cloned version
instead lets us avoid the memcpy.

PiperOrigin-RevId: 208710603