One of your tails is at Y=5.0 -- which creates a crossing of panels between your horizontal and vertical tails. Moving it to 5.1 to match the other tail and booms helped a lot. Also, adding some panels to the wing is a reasonable thing to do.
One of your wake lines is going unstable and causing the asymmetry and blowing up the solution.
My modified model (shifting tail, adding some wing resolution) held together to about 7 deg alpha. It blew up from there -- in this case, if you look closely, you can see a wing tip wake line going through one of the tail surfaces.
In this case, going to a Panel method actually helped quite a bit. I threw some more resolution at the pod in the middle this time -- and only ran half as many points to speed things up, but it held together rather well.
Rob