redo the planck likelihood without requiring Planck data

[qyx268]

Currently, to use the Planck PS likelihoods, there are external data and code (e.g. clik) needed. A new python package has been published to do this. Perhaps, it's a good idea to walk away from the Planck code and to use this planck-low-py, in particular when PR #39 is still not merged...
paper: https://arxiv.org/abs/2104.05715
code: https://github.com/heatherprince/planck-low-py

[steven-murray]

I'm all for making things more python-native, but I'm a bit worried that that code is not yet ready for production. Is there perhaps a way we could help bring it up to production level?