[DesignTools] makeBlackbox() to use phys netlist not logical

[eddieh-xlnx]

Currently, makeBlackbox() leverages the EDIFNetlist to find all leaf cells inside a particular point of the hierarchy. In the case of encrypted or detached netlists, this is not possible. Instead, this PR proposes that only the physical netlist is used by assuming that any Cell that starts with the hierarchical name must be within the hierarchy to be blackboxed.

Am I missing a reason why this shouldn't be done? The only reason I can think of is that simply relying on the Cell name prefix may be imprecise, but is that enough of a (rare) reason to not proceed?

This PR also:

• Fixes a situation where the intra-site routing of FF routethru-s were not being unrouted correctly
• Depends on a new feature (in a future release) to defer the unrouting of sink pins such that they can be batched together and performed more efficiently later