Incorrect calculation of maneuver time of flight

[chrisbor]

Hi!
I think there is a bug in the calculation of a maneuver's tof. The method get_total_time() returns the sum over all the maneuver's _dts. However, if I understand it correctly, the first element in _dts for bielliptic and hohmann transfers is not an actual flight time but only the epoch of the initial orbit's pericenter which the orbit has been back-propagated to (and, hence, the maneuver and other _dts elements start at).
For a few cases, I've double-checked the return values of get_total_time() using the formulae from the documentation and they seem to indeed be wrong.

Hope this info suffices - it's my very first time using GitHub :)