Allow for both linear and circular plasmids, as well as regions

[babayagaofficial]

(although regions with caveats: assume embedded into circular genome, and that they are all oriented 'correctly', as two regions that are the same, but one is the reverse of the other, will have DCJ distance 1 under current assumptions)

addresses issues #44 and #6