Smallest set of smallest rings inconsistent for fused rings

[rwest]

There's an element of chance depending on which atom you choose first.

[rwest]

This in turn messes up the "is it in more than one cycle" method, which then makes the thermo random, see #155

[vrlambert]

I pushed a commit which should fix this. I implemented the changes we talked about at the RMG study group a few weeks ago where it removes first the root node and then other nodes with only one edge.

90866a5

[rwest]

Closed by 90866a5.
Thanks @vrlambert