Correct Writing of Steps to simfile.dat

[fjclark]

Hello,

Currently, the step number written to simfile.dat seems to be too low by energy_frequency. This results in a final step number energy_frequency lower than expected, and an initial step number of zero, despite no data being stored until after the first energy_frequency steps. I've offset the steps written by simfile.dat by energy_frequency to account for this.

Thanks, Finlay

[chryswoods]

This looks sensible to me :-)

I'll leave it to @jmichel80 to merge, as it affects the OpenMM code.