Fix and document clearsky_model for ModelChain

[cwhanse]

• Closes #xxxx
• I am familiar with the contributing guidelines
• Tests added
• Updates entries in docs/sphinx/source/reference for API changes.
• Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
• New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• Pull request is nearly complete and ready for detailed review.
• Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

Clarifying why ModelChain has attribute clearsky_model. See discussion.

[kandersolar]

Can someone point me to a line of code where clearsky_model actually has an effect? From what I can tell, it is not used anywhere. The clear-sky fallback was removed in v0.7.0; maybe this parameter should have been removed as well but was overlooked? See:

• Remove clear sky fallback option from ModelChain.prepare_inputs and run_model #705

[cwhanse]

Can someone point me to a line of code where clearsky_model actually has an effect?

Now that I look carefully, I can't. I saw the line in modelchain._complete_irradiance where a clear-sky model is run and thought "Aha" and didn't look further.

Rather than drop clearsky_model (I'm OK with that), we could instead pass model=clearsky_model.

[kandersolar]

LGTM