High: PE: Bug N-191176 - Implement interleaved ordering for clone-to-clone scenarios
This is important because it affects recovery of stacked clones.
Previously the recovery of a clone instance caused the recovery of every
instance of the clone sitting on top of it - which far far from useful.

Previously the lower resource had to be finished starting before the
"upper" one could begin to (and the inverse for stops), this requirement has
been relaxed for interleaved clones and the "upper" one can begin starting on
nodeX as soon as the "lower" one has finished starting there.

See the change to the inc7.dot file for an illustration

Mercurial revision: 4c5c6e8390570ff6e7e543ae8be1c96341565e3d