Add helper functionality to plot quantities against temperature with axes labeled in both K and eV

[namurphy]

Feature description

It would be helpful to have a plotting helper function that would let us have one x-axis or y-axis show temperature in kelvin and another showing $k_BT$ in eV on the opposite axis.

Motivation

Different fields in and adjacent to plasma science have different conventions for representing temperature. Fields like solar physics and astronomy often use K or MK, while laboratory plasma scientists tend to represent temperature (times an invisible Boltzmann constant) in eV or keV. In a cross-disciplinary meeting, it can take mental bandwidth (brainwidth? 🧠) for some folks to convert temperatures in K to eV, or from eV to K. Having an easy way to plot quantities against temperature with labels in both eV and K would help make plots that can be better understood across communities.

Implementation strategy

• I'm not sure where this would be best implemented. It could possibly go in plasmapy.utils, or we might want to include
• We'd need to make sure that this would handle linear, linear-log, and log-log plots.
• We'd probably want to use the astropy.visualization.quantity_support context manager, as well as the astropy.units.temperature_energy equivalency.
• It'd help to have an optional particle keyword argument so that the symbol of the particle can be included in the labels (i.e., $T_e$). To do this we'd probably need to add an extra attribute on Particle.
• We'd need to make sure that this would work in multi-part plots.
• We might want to add an admonition to the docs to say that we should not have the Boltzmann constant be silent, i.e. that we should say $k_B T$ instead of just $T$ for temperatures in eV.
• We probably want to include a way to specify the specific units (i.e., K vs. eV or MK vs. keV).
• How to test matplotlib plots is discussed in matplotlib's testing guide.
• It'll also be helpful to include an example Jupyter notebook to demonstrate this, or to modify an existing notebook. It may be better to include this as a separate, follow-up pull request.

[kuchtact]

I think that having a more general function where any two variables can be passed and the conversion is taken care of could also be useful. As an example doing meters and ion skin depth.

[pheuer]

I think that having a more general function where any two variables can be passed and the conversion is taken care of could also be useful. As an example doing meters and ion skin depth.

Really it would be nice if matplotlib just had a better way of adding a second axis with a fixed relationship to the coordinates on the other axis

Current way to do this in matplotlib? https://stackoverflow.com/questions/55907892/matplotlib-secondary-axis-with-values-mapped-from-primary-axis/55909132#55909132