Fix empty reference location

[kosack]

This adds a fallback dummy location to subarray.reference_location if the metadata in the simtel file doesn't exist. The dummy location is on "null island" (lat=0, lon=0), but at the correct height as read from the simulation_config.

This was a problem for pre-Prod6 files, which didn't contain the metadata necessary to determine the reference location.

[Tobychev]

Would it be possible to add a test that checks "proper" initialisation when metadata is missing?

[kosack]

This makes me realize that we don't have a test that checks the interface of all installed EventSources. It's not so easy to do though, as you would need the right inputs for each one. But it would be good to have a way to verify that an EventSource plugin creates all the necessary attributes

[maxnoe]

Any particular reason for this move?

[kosack]

Yes, it needs to come after parse_simulation_header, otherwise I can't use the SimulationConfigurationContainer

[maxnoe]

Docs failure is fixed in main, please rebase

[maxnoe]

One of the example notebooks no fails because it tries to print a subarray without a reference location:

WARNING: /home/runner/work/ctapipe/ctapipe/examples/core/InstrumentDescription.py failed to execute correctly: Traceback (most recent call last):
  File "/home/runner/work/ctapipe/ctapipe/examples/core/InstrumentDescription.py", line 49, in <module>
    newsub.info()
  File "/home/runner/work/ctapipe/ctapipe/ctapipe/instrument/subarray.py", line 129, in info
    printer(f"Height   : {self.reference_location.geodetic.height:.2f}")
AttributeError: 'NoneType' object has no attribute 'geodetic'

[kosack]

One of the example notebooks no fails because it tries to print a subarray without a reference location:

Seems it was select_subarray() not propagating the reference location. Fixed. Also, I now changed SubarrayDescription's constructor to require the reference location (and telescope info), to avoid that it is ever set incorrectly.