Questionable results when using the adhoc solver #545

I've had reports from the MEng students that supply values do not match consumption values when using the adhoc solver (UK Transport model). Switching to the scipy solver fixes the problem.

I have no idea what the problem is, just making a note of this as a warning to anyone that's still using the adhoc solver.

Ideally I'd quite like to drop the adhoc solver, although we can't do this yet as the scipy solver is unsuitable for some models due to memory issues. I think my efforts are probably better spent improving the scipy solver so that the adhoc solver becomes redundant, rather than fixing bugs in the adhoc solver (unless this turns out to be really easy to fix)