Fix issue #371

[lohedges]

This PR closes #371 by ensuring that the output of Connectivity::findPath is reproducible and considering all 4 atom paths when adding missing dihedrals for when looping over 1-4 pairs. I have tested this via the SOMD1 compatibility layer in SOMD2, but can try to come up with a unit test here if desired.

• 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 added a changelog entry to the changelog (we will add a link to this PR as part of the review): [y]
• I confirm that I have permission to release this code under the GPL3 license: [y]

[lohedges]

Oh, and just to add that I've checked that the legacy AmberParams tests still pass.

[chryswoods]

Looks good - I think the only thing to keep an eye on is if the cost of finding all paths becomes very high for molecules with lots of rings?

[lohedges]

Yes, good point. I'll get users to test. Hopefully we can run it over a range of inputs during the somd2 tests to see how it performs.

[lohedges]

Actually, it will make no difference since findPath implicitly calls findPaths anyway. The only additional overhead will be from sorting the paths, but this should be (hopefully) offset by the fact that we now have an early exit condition for the dihedral addition loop, i.e. we don't add the same dihedral multiple times. In any case, we will see how things go once we've run a few benchmarks.