The specialized software for parabolic antennas is TICRA, from Denmark. I haven't used it but I know that this is their specialty.
If you would like a general purpose solver which is also good for parabolic antennas, then you could try MoM solvers.
WIPL-D can solve very large parabolic antennas using rigorous higher order MoM, so this can be a good option. You should try it out.
When using FEKO, you would probably have to go to a hybrid MoM/PO solution for larger parabolic antennas, and would probably able to solve the smallest ones by pure MoM.