Update monitor diagnostics

[schlunma]

Description

This PR updates the monitoring diagnostic.

• Update the default plot path in the documentation so that is reflects the true default path (I didn't change the actual path) and allow the usage of the tag {plot_dir} (default ESMValTool plot directory) in the recipe.
• Always use the user-defined output-file-type for all plots.
• Added automatic rasterization for map plots that can be turned off using the option rasterize_maps=False (default True). Note that I had to implement this via Artist.set_rasterized(True) since it's not possible to pass arbitrary keyword arguments (and thus rasterized=True) to the plotting function in mapgenerator.
• Added captions to the provenance record of all plots.
• For timeseries plots, I think the original intention was to also include a smoothed version of the timeseries in the same plot for longer time series. However, the way it was coded, only the smoothed version was plotted, since the plot is saved and closed after each call of plot_cube. I changed that so that the smoothed version gets plotted in a separate plot. Feel free to suggest any alternatives, but note that maybe an update of mapgenerator is necessary for that.
• Adapted some default options for suptitle positions because the titles were sometimes overlapping, especially for stereographic plots.

• Closes Format of output plots in recipe_monitor.yml is sometimes hardcoded instead of configurable #2561
• Link to documentation: https://esmvaltool--2608.org.readthedocs.build/en/2608/recipes/recipe_monitor.html

---

Before you get started

• ☝ Create an issue to discuss what you are going to do

Checklist

It is the responsibility of the author to make sure the pull request is ready to review. The icons indicate whether the item will be subject to the 🛠 Technical or 🧪 Scientific review.

• 🛠 This pull request has a descriptive title
• 🛠 Code is written according to the code quality guidelines
• 🛠 Documentation is available
• 🛠 Tests run successfully
• 🛠 The list of authors is up to date
• 🛠 Any changed dependencies have been added or removed correctly
• 🛠 All checks below this pull request were successful

New or updated recipe/diagnostic

• 🧪 Recipe runs successfully
• 🧪 Recipe is well documented
• 🧪 Figure(s) and data look as expected from literature
• 🛠 Provenance information has been added

---

To help with the number of pull requests:

• 🙏 We kindly ask you to review two other open pull requests in this repository

[schlunma]

Could anyone take a look at this? It would be nice to merge this soon since I have a follow-up PR that extends the monitoring diagnostic with an additional plot type.

[sloosvel]

apologies, but I am a bit packed with another couple of diags! If you are in a hurry feel free to ask someone else to review, else I'll take a look as soon as I am done with those tasks.

[schlunma]

No worries, just wanted to make sure that we don't forget about this!

[sloosvel]

Taking a look on this on monday! sorry for the delay

[sloosvel]

For timeseries plots, I think the original intention was to also include a smoothed version of the timeseries in the same plot for longer time series. However, the way it was coded, only the smoothed version was plotted, since the plot is saved and closed after each call of plot_cube. I changed that so that the smoothed version gets plotted in a separate plot. Feel free to suggest any alternatives, but note that maybe an update of mapgenerator is necessary for that.

Having the smoothed versions in a separate plot is fine for me. I also spoke to the maintainer of mapgenerator, if any of you wants to contribute changes for the monitor diags into the package feel free to do so. Just don't expect a lot of involvement to keep up with ESMValTool needs because it's an internal package that is mostly used by other groups in the dept, but if you feel like contributing go for it.

[sloosvel]

Overall the changes look fine to me and do not break anything. I could run the recipe setting several output files types without any kind of issue. As mentioned, if you want to contribute the plot changes to mapgenerator you can, but it's not necessary.

[schlunma]

Thanks for the review Saskia! 🎉

As mentioned, if you want to contribute the plot changes to mapgenerator you can, but it's not necessary.

I agree, that shouldn't be necessary.