Add naming rule for instance variables

[chjaesch]

Regarding to the PEP8 namely conventions, method names and instance variables shall pass the function naming rules (lowercase with words separated by underscore): https://peps.python.org/pep-0008/#method-names-and-instance-variables
Therefore, it would be beneficial to regard this in pep8-naming by adding an naming rule for instance variables, similar to N806.

For example, self._fooVariable in

class AClass:

    def __init__(self) -> None:
        self._fooVariable = 42

should be reported.

[sigmavirus24]

I believe this has been discussed in depth before. The short answer as to why this is hard is that subclassing something using camelcase variables then will cause pep8-naming to flag that usage as well as it's indistinguishable from the example you gave. The rate of false positives would be too high and we don't execute code to be able to determine if an instance attribute is defined for the first time in the scope we see it or if it comes from a parent

[chjaesch]

Regarding PEP8 conventions, using CamelCase for variables in a parent class would also not be correct.
Therefore, it would be consistent to mark it as an error if it were used in a child class. If it is implemented as an individual naming rule, the user can suppress this if necessary.
Otherwise, the non-use of "lower_case_with_underscores" will not be sanctioned and will continue in the future.