Object oriented mode solver about femwell

[mdecea]

This looks great, I think it will make it much easier.

I would add two requests as someone who has gone through the process of starting to use femwell recently:

1. Please, make variable names a bit more understandable.
   In particular, I find the standard syntax for finding the modes very confusing:

lams, basis, xs = compute_modes(
    basis0, epsilon, wavelength=wavelength, num_modes=nmodes, order=1
)

xs is in reality E fields, but with this names it looks like some kind of coordinates. lams are effective indices but with this name look like wavelength.
My suggestion, rename xs --> E_fields, lams --> complex_n or something similar.

2. I think it would be extremely useful to have an initial doc explaining a bit the terminology for someone who is not super familiar with FEM methods. What is a basis, what is an element, what does it mean to "construct" a basis with certain selected elements...

As an extra, documenting the methods would also be nice :).

Happy to help!

[HelgeGehring]

Thanks for the feedback!

1. I think this will be fixed already by the object oriented refactoring, right? compute modes would only return a list of modes (or an object which can be treated like a list, but has extra functions like sort). The names of the fields are listed above, I think then things should get very clear :)

2. Makes sense! let's add something soon :)