Improvement: Use other quality evaluation in EigenGraspResult

[JenniferBuehler]

So far, the energy of the simulated annealing search is passed on as grasp quality value. It is worth investigating if it's possible to compute other quality values.

This issue may be related to issue 1: Problems were encountered with using quality evaluation during the search. Both issues may be resolved in the same process.

It seems that during the search, this may be a bit of work with the original graspit code. See also this page in the GraspIt! manual.
However it could be possible to re-evaluate the results after planning, maybe after applying an AutoGrasp. The results could then be sorted according to the new quality measure. Repeated application of the planning algorithm could help to get more different results (due to the randomized search process).