GitLab

Refactored ForceImpl, lots of KernelImpl subclasses, and other related classes to avoid redundant calculations when requesting a State with both forces and energies

Fixed unit conversion bug in Monte Carlo barostat where pV work was being divided by a factor of AVOGADRO*1e-25 (kJ/mol)/(bar nm^3), rather than multiplied by it.
Checking this bug in since Peter Eastman is on vacation; he should double-check the fix upon return.

Use a consistent (and more accurate) value for electric constant everywhere.  Also cleaned up handling of constants and units in various places.

Added a version of Platform::getPlatform() that looks up platforms by name.  Added a Context constructor that allows platform-specific properties to be specified.

Redesigned CustomNonbondedForce to eliminate combining rules and allow the expression to directly use parameters for individual particles.

Free energy plugin added
Plugin will not run w/ Obc or GB/VI forces unless line 2004 of gpu.cpp (gpu->sim.totalNonbondOutputBuffers  = 2*gpu->sim.nonbondOutputBuffers;) is commented in -- working on removing this constraint
Also unit tests for GB/VI currently fail