Unbound energy

[klgunst]

• When using VINA or VINARDO as scoring and using refinement of the structures after docking, the Unbound System's energy was obtained from the worst pose instead of the best pose.
  I couldn't reproduce it locally, but I expect it to close issue python script results in unexpected very negative affinity score #68.
• Fix typo as pointed out in issue python script results in unexpected very negative affinity score #68 and fixed sign-inconsistency for unbound system's energy between different scorings.
• Expose score(unbound_energy) in python bindings and CLI (through --unbound_energy when using --score_only). This is handy when you want to rescore a pose with a 'custom' unbound energy (i.e. if you are not using the intramolecular energy of that particular pose as unbound energy).

[diogomart]

Thank you @klgunst for this PR. I can confirm that the large energies reported in #68 no longer occur. I'll run a few more checks and will probably merge soon.