Fix/leapr array

[whaeck]

@HunterBelanger Everything looks good now. Tests pass so we're good to go. The failing tests must have been fixed in the last update you made.

[kahlerac]

Surprising to not see any deallocation statements in subroutine contin. contin is called from within a loop over user specified temperature. Do existing tests include a multiple temperature leapr job to assure we don't get an already allocated runtime error?

[HunterBelanger]

Test 23 should have multiple temperatures in the LEAPR input. I believe that since Fortran 95, allocatable arrays are automatically deallocated when they go out of scope. I think this is why there are no deallocation statements at the end of contin. I didn't see any for the other arrays like p, tlast, tnow, and xa, so I didn't add one for maxt.

[whaeck]

Deallocation will be required when you run LEAPR multiple times in the same input deck. This is an issue we have seen in the past with other modules like ERRORR. In this case, I don't think it is necessary to protect users against themselves.

As such, I'll pull this into develop. If the deallocation comes up later, we'll just fix it then.