Rework or clarify the way we use settings

[SMoraisAnsys]

There seem to be some misunderstanding with how the code base works and how we want the final result to be.

This observation comes from @Alberto-DM who noticed that even when testing and setting the pyaedt.settings to not log into a file, we end up with a file being written. This comes from how python works. For example, in the following code extracted from our conftest.py file

from pyaedt.generic.settings import settings

settings.enable_local_log_file = False
settings.enable_global_log_file = False
settings.number_of_grpc_api_retries = 6
settings.retry_n_times_time_interval = 0.5
settings.enable_error_handler = False
settings.enable_desktop_logs = False
settings.desktop_launch_timeout = 180
settings.release_on_exception = False

To import settings we are loading pyaedt.__init__.py, then pyaedt.generic.__init__.py and then pyaedt.generic.settings.py.
However, doing so leads instantiating settings = Settings() since other modules loaded through pyaedt.__init__.py do so.
Since the default value of enable_global_log_file is True then, if for whatever reason one of the file triggers some logging - like logging that one has not pyvista, numpy, or whateer - we'll have a file with the logged content, while the user might think that by setting settings.enable_global_log_file = False it prevented it.

I'm not sure of the correct approach to handle this as making changes might break users code or might require deep changes.

I'll investigate on the coming days and let you know what approach I can think about. Of course, any feedback is much appreciated !

[Samuelopez-ansys]

@SMoraisAnsys Could we close this issue right?

[SMoraisAnsys]

Closed by #5092