Keep inline annotations over docstring generated ones

[OriolAbril]

I have started playing with this a bit in between meetings.
The other place with the "Replacing inline type hint" message wasn't actually being triggered,
I am not sure if it has to do with the condition not being correct or with something else,
I managed to do something similar to the return method in the param method which seems to work.

For now I have updated but kept the messages, however, I think they should not be present by default,
maybe only when requested through a verbosity flag or something.

I also noticed that there are some import related updates. Should these be modified or simply
skipped when keeping the inline type-hint?

Release note

Ensure that existing inline annotations aren't overridden by annotations in
docstrings. This allows documenting types inside docstrings that are not
(yet) supported by Python's type system. Instead, more general inlined type
annotations can be used as a fallback.
Previously, annotations in docstrings would take precedence.
{label="enhancement"}

[lagru]

Thanks @OriolAbril. I finally got around to reviewing this. I added a few more updates and tweaks. E.g. I also made sure that assignments / attributes behave the same as parameter and return annotations.

[lagru]

Let me know if you want to give it a look. But I think this should work for you. :)

[OriolAbril]

Thanks! I have done some quick checks and found no issues. Also as the warnings don't fail the respective local pre-commit check I will be able to start adding it to CI without problem from 0.4 onwards 🎉 (I'll probably be back with other issues, but I can always add an inline override as a hotfix now)

[lagru]

Thanks @OriolAbril for getting this going!