Code to do inference from line intensity maps

TODO:

• Implement map_object saved with header (like real observed temperature map)
• Automatically run on same input data as previous run
• Automatically run on same experiment and mcmc setup as previous run
• Continue from state of previous run
• Lots of ideas for different models to compare

In progress:

• Tests, and automatic tests (pytest?)
• Test parallelization with multiple hosts etc. benchmarking
• Make run_mcmc save results and parameters used together with

Done:

• Save imput data in log
• Sampling initial points from priors
• Create function to automatically generate mock data
• Add instrumental noise
• Log file could summirize results (e.g. theta = 2.3 +- 0.21)
• Let run_mcmc be run with parameter files as command line arguments instead to use them instead of the default ones
• Add instrumental beam
• Implement cosmology (angles, redshift, Mpc etc.) (Check and fix this) - Decide on consitent way of defining, dx, dy, dz, v_vox and v_full
• Include additional model output (e.g. luminosity functions)
• log file - save experiment params as well - save exact command used when running current script in log file (e.g. python run_mcmc.py experiment_params_comap_lignt.py mcmc_params_full_parallell.py) (not easy to do)
• Implement functions to calculate covariance matrices