irradiance.py updates: glossary term links and units

[RDaxini]

• Relates to: Standarizing pvlib typesetting of units - standarization for common variables? #2205, [DOC]: params & units guidelines in contributing page #2248, Glossary page updates tracking #2284, GSoC feedback on the pvlib contribution process #2081
• 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.

Updating the units to superscript and linking key terms to their glossary page definitions.
Follow up PR(s): add some of these key terms into the glossary, enhance existing glossary term definitions (units and explanation)
Note: we can now view definition tooltips by hovering the cursor over the linked glossary term (context: #2290)

[RDaxini]

I could modify the scope of this PR if we want to see some of the changes implemented in 11.2. I don't think these changes are urgent though so I'm also happy to keep working on it and merge the completed version in 11.3. Not sure what reviewers would prefer though--- on second thoughts many small revisions covering the entire irradiance.py might be a pain to review. Thoughts @kandersolar @AdamRJensen @cwhanse ?

[RDaxini]

I went for the format the seems to have the most in favour and is what is currently outlined in our contributing guidelines: link to nomenclature, period, units, no period

All parameters that have an entry on the nomenclature page and require a unit should now have both. The parameter description structure for all parameters should also now be consistent.

[RDaxini]

@cwhanse thank you for the meticulous review!

[RDaxini]

Good to go, I think. Any further feedback?

[IoannisSifnaios]

One question from my side: why in some functions you haven't written down the limits, while in some others you have?

E.g., in pvlib.irradiance.gti_dirint the zenith and azimuth are:

While in pvlib.irradiance.perez_driesse:

[RDaxini]

@IoannisSifnaios good point, thanks. I think it's just my error. I don't think any of these functions enforce the limit. I feel like there was a discussion somewhere else about whether these limits should be included in docstrings but I cannot remember where... @cwhanse might have been a part of it, can you advise?

[cwhanse]

Value limits in docstrings: much of that text was ported to python from the original Matlab code. There have been two, sometimes competing objectives for value limits: 1) describe the range of values that the code can process, and 2) describe limits that are reasonable for the physical device or process being modeled. Violating #1 would give an error or an incorrect result; violating #2 gives a value that's not meaningful (that's not the same as being incorrect). For example, the irradiance.aoi function works with any value of its input angles but only returns meaningful results for combinations of surface_tilt, surface_azimuth and solar position.

It is my opinion that only code-imposed value limits should be present in the definition of a parameter. If there are also reasonable ranges of values, I would put those in a Note section.

For definitions in the Glossary, I don't think we should include either type of range, since they probably depend on the use of that quantity in a function.

[RDaxini]

Thanks @cwhanse
I'll update this PR soon. I haven't disappeared, just been busy lately.

It is my opinion that only code-imposed value limits should be present in the definition of a parameter. If there are also reasonable ranges of values, I would put those in a Note section.

+1