x_orientation Stuff

[r-pascua]

We've recently had some hiccups on the Validation side of things relating to the x_orientation attribute. For visibility simulations, the x_orientation should match whatever it is in the beam, assuming it is defined. That said, considering this repo is focused on simulating HERA data, we should consider whether we enforce a convention that is consistent with actual HERA data.