Fix issue #76

[lohedges]

This PR closes #76.

(Just to note that there is still quite a lot of code duplication in the BioSimSpace.Process._openmm.py module since it has not yet been ported to the new BioSimSpace.Config configuration generator sub-package. This is because we write a full Python script, not a configuration file, so it doesn't quite fit the existing logic. I do plan on reworking things to make the code generation more modular, particularly now that the use of mixins means that things like restraints are supported by all protocols. I originally separated the protocols so that it was easier for a user to find a read the exact specification, and for us to tweak them independently while figuring out what worked best.)

• I confirm that I have merged the latest version of devel into this branch before issuing this pull request (e.g. by running git pull origin devel): [y]
• I confirm that I have permission to release this code under the GPL3 license: [y]

Suggested reviewers:

@chryswoods

[chryswoods]

Looks good. I think you are right to duplicate code and review things later. The mixin code should make this relatively easy.

[lohedges]

I've now fixed the issue that this workaround solves in my fork of openmm here. I'll see what they say in the issue thread that I opened and raise a PR if they're happy. I'm not sure how long it would take to work it's way into a release, but it would be easy for me to revert the changes introduced in this PR at that point. This might be desirable if people want to be able to run the script in a minimal environment, e.g. without parmed, which isn't a dependency of openmm itself.