Deprecate parse_psm3 and parse_cams

[kandersolar]

• Progress towards Consider removing the parse_ iotools functions #2444
• 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.

To keep reviews manageable, I thought it better to not deprecate all the parse_ functions in one PR. This one does just PSM3 and CAMS, which are both straightforward.

[AdamRJensen]

LGTM!

I really think this is a good move, and will reduce maintenance burden in the future.

[echedey-ls]

Absolute LGTM. Two minor nitpicks down below, feel free to ignore one of them (at most!).
Just as a side note, I prefer setting the versions at which they will be removed from the start, but it's not a strong opinion.

[echedey-ls]

Suggested change

.. versionchanged:: 0.9.0

May be time to clean-up this admonition

[kandersolar]

+1, but I suggest we do it as part of a wider survey of old versionchangeds that could be removed

[kandersolar]

Just as a side note, I prefer setting the versions at which they will be removed from the start, but it's not a strong opinion.

I know. The argument in favor of leaving it unspecified is that it gives us flexibility to make the *.0 releases sooner than we expected, and I think the length of time is more important than the version numbers. (I know you know this, I am just explaining so that you don't think there is some better reason.)

[echedey-ls]

Just as a side note, I prefer setting the versions at which they will be removed from the start, but it's not a strong opinion.

I know. The argument in favor of leaving it unspecified is that it gives us flexibility to make the *.0 releases sooner than we expected, and I think the length of time is more important than the version numbers. (I know you know this, I am just explaining so that you don't think there is some better reason.)

Explanations are always welcome. Thanks @kandersolar .