v3: Make dt mandatory in documentation

[BenjaminRodenberg]

• Merge Update (some) breaking API changes #299

[BenjaminRodenberg]

This implements the changes from precice/precice#1622 in the documentation.

[fsimonis]

There are some contextual changes that are missing. Especially the way we handle the max time step size has changed.

[uekerman]

I don't think that this is a good strategy (see also above): To tell users first how to do it the wrong way, and then the right way. Maybe we need to re-order our steps?

[BenjaminRodenberg]

Now that the waveform API is not experimental anymore but the default, let's put some information from here into "Step 5 - Non-matching time step sizes". I think this section needs some updates and refinement anyway. Additionally, I think that the topics that we currently explain in "Step 5 - Non-matching time step sizes" are definitely more advanced and exotic than some of the basic topics that we explain here. So yes: let's reorder some things.

We should still keep the more advanced background information (interpolation degree, higher-order time stepping) here to not provide too much information too early.

[BenjaminRodenberg]

There is a $\tau_n$ here. The point about the $\tau$ is the following: Actually it can refer to any point in time. So $\tau$ could be $t_n$, if we are at the beginning of the window, but also any arbitrary point in time between $t_n$ and $t_{n+1}$.

[uekerman]

I didn't get this from the picture, seems like you use $t_n$ and $\tau_n$ for the same thing.
Above you write

... relatively to the beginning of the current time step starting at $\tau_n$

This should be $t_n$, no?

[fsimonis]

Some of these pages also still refer to waveform-order instead of waveform-degree.

[uekerman]

I would suggest to merge here and improve later.