The tests for the TATT intrinsic alignment systematic should be refactored

[marcpaterno]

Problem description

Several of the tests for the TATT intrinsic alignment systematic are very long (so long that pylint was complaining about them). These complaints are currently silenced in the code.

Since the complaints are signs that the tests really are too complex (and thus it is hard to know they are really verifying what needs to be verified), they should be refactored and the commands to silence the complaints should be removed.