add reader for Cobaya's MCMC chains (requires GetDist)

[lukashergt]

This introduces a chain reader for Cobaya's MCMC chains. Cobaya (different from CosmoMC) no longer provides .paramnames files for its MCMC chains, but instead works with .yaml files which are simultaneously Cobaya's input files. These contain all information on the sampling parameters and derived parameters, including the tex string. The CobayaMCMCReader that is implemented here uses GetDist to infer the parameter labels used for the pandas dataframe as well as the tex strings for the .tex dictionary. Since GetDist is a requirement for Cobaya anyhow, having it as a requirement for anesthetic's CobayaMCMCReader shouldn't be too bad.

Fixes #129

Checklist:

• I have performed a self-review of my own code
• My code is PEP8 compliant (flake8 anesthetic tests)
• My code contains compliant docstrings (pydocstyle --convention=numpy anesthetic)
• New and existing unit tests pass locally with my changes (python -m pytest)
• I have added tests that prove my fix is effective or that my feature works

[codecov]

Codecov Report

Merging #167 (b83488c) into master (0b37b64) will increase coverage by 0.00%.
The diff coverage is 95.74%.

@@           Coverage Diff           @@
##           master     #167   +/-   ##
=======================================
  Coverage   95.09%   95.10%           
=======================================
  Files          16       17    +1     
  Lines        1672     1715   +43     
=======================================
+ Hits         1590     1631   +41     
- Misses         82       84    +2     

Impacted Files	Coverage Δ
anesthetic/read/getdistreader.py	92.42% <ø> (-0.33%)	⬇️
anesthetic/read/cobayamcmcreader.py	95.23% <95.23%> (ø)
anesthetic/read/samplereader.py	78.57% <100.00%> (+3.57%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0b37b64...b83488c. Read the comment docs.

[williamjameshandley]

This looks like a very nice addition. I assume that since you're a Cobaya aficionado you've tested this 'in the wild' in addition to the tests you've put in here?

Once @Stefan-Heimersheim has finished #158 for dynesty, I think the only one we're missing is cosmosis. @Pablo-Lemos does cosmosis have it's own chain formatting, or does it just rely on polychord/multinest?

[Pablo-Lemos]

I believe CosmoSIS has its own formatting, and in fact annoyingly does not generate a paramnames file