[fix] Change solver interval to align with the evolution time

[maxbriel]

Related to issue #794.

The time from the solver was not correctly added to the binary.time.
The solver here runs from 0 to max_time-binary.time.
This solution adds the final solver time to the binary.time

Alternative solution is to change the solver from binary.time -> max-time. This would be more in-line with the normal detached step.

[sgossage]

I think what's going on is actually that the CO step solver uses a bad time interval. Right now it always starts at 0 and spans up to the max time minus the current binary.time. It should start at the current binary.time and go until the max time (similar to how it is done for normal stars). I suggest the following change.

Right now we have:

        t0 = binary.time
        ...

        while status == -1 and n < 6:
            try:
                res = solve_ivp(self.evo,
                                events=self.evo.ev_contact,
                                method="BDF",
                            t_span=(0, self.max_time - t0),
                            ...)

            ....
            t0 += res.t[-1]

and we should use

        t0 = binary.time
        ...

        while status == -1 and n < 6:
            try:
                res = solve_ivp(self.evo,
                                events=self.evo.ev_contact,
                                method="BDF",
                            t_span=(t0, self.max_time),
                            ...)

            ....
            t0 += res.t[-1]

[maxbriel]

That's good. Either we change the interval or how we add the time. I'm happy to go with the solver change.