Ability to inject initial conditions (only density field)

[jordanflitter]

This PR implements two new more features.

The first one is the ability to inject hires_density_array to p21c.compute_initial_conditions. If a non-zero array is injected, then the usual Gaussian sampling in Fourier space is skipped, and then the rest of the computation continues. As before, this modified initial_conditions can then be used as an input to run_coeval (or run_lightcone) in order to make a "roundtrip test", something that could be useful for @nikos-triantafyllou's project.

The second new feature is the ability to run_global_evolution with delta_z0 which is not necessarily 0. This could be helpful in making (speedy) linear perturbation theory* with 21cmFAST.

*Note that this is a scale-independent perturbation theory, since we currently assume that the density field evolves in a scale-independent manner (as we simply multiply it by the scale-independent growth factor). This is actually not true in the dark ages, where the baryon density field evolves in a scale-dependent manner (unlike CDM, which roughly evolves in a scale-dependent manner), but this will wait for another PR

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.16%. Comparing base (943ca5f) to head (38582c8).
⚠️ Report is 14 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #613      +/-   ##
==========================================
+ Coverage   88.13%   88.16%   +0.03%     
==========================================
  Files          32       32              
  Lines        4729     4741      +12     
  Branches      796      800       +4     
==========================================
+ Hits         4168     4180      +12     
  Misses        401      401              
  Partials      160      160              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[steven-murray]

Thanks @jordanflitter this is almost there, we discussed just now the following:

• Changing the name of the hires_density_array parameter
• Changing the with_value_not_computed to be private
• Updating compute_xray_source_field to use your approach
• Adding your paper (and others) to show_references()
• Adding a small section to a new tutorial that promotes your new feature.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[jordanflitter]

Thanks @steven-murray for your comments! I renamed some of the variable + function names, and I also updated the tutorials and the faq to give more information on the new features, as we discussed.

Unfortunately, we cannot at the moment get rid of the allow_already_computed hack in outputs.py and use "my approach" in compute_xray_source_field, I gave some details in #552.

As for your last two comments (include my paper in show_references() and promote it in the docs), I'll do it as part of #534.