v3.0: Async? + API feedback

[bain3]

I am thinking about 3.0, and if the library should be async, or not. What are your thoughts? React with 👍 (async), or 👎 (sync), please.

It really depends on what people are using it for, and if it would make it easier to use.

If you think the API should be changed, or have some feedback, leave a comment :)

[bain3]

What I have so far:

• clean up parsing, split dataClasses file, since it is becoming a mess
• try to use the dataclasses module.
• make all the properties (like Period.grades) functions, since they actually make requests, which is kinda hidden
• don't calculate averages if there are none, just stick to what pronote gives us
• keep communication logic, login, etc... inside the communication class. The client should really just make requests, not caring about encryption, or how to log in.

[tarneaux]

I'd also add a general code cleanup.
For example, why do both teacher_name and teacher_names exist for the same pronotepy.lesson dataclass? Shouldn't one of them be removed? Is Lesson.virtual_classrooms still relevant (does pronote still have that feature)?

[bain3]

Yes, definitely, that's the main point of v3. The teacher_name attribute is for backwards compatibility. There are multiple things that should be removed, but were left there, so that we don't break the API unnecessarily.