better way to do datafiles/noisefiles in likelihood

[steven-murray]

At the moment, the likelihoods all expect that the simulate() will return a list of dicts, which accounts for the case where you might have several co-eval simulations at different redshifts (or different lightcone chunks), where the data should be saved in different files. However, this makes for a bit of awkwardness when writing likelihoods that have no need of multiple files (i.e. everything has to be zero-indexed before use).

This should be dealt with consistently to provide a nicer way to write likelihoods.

[BradGreig]

It is not obvious to me at this point how one would deal with this. I reckon this might become obvious once we start playing around with it on larger-scales for different applications.

[steven-murray]

Closing due to it moving to the above referenced issue.