Hi Jaime,
I am one of the maintainers of pyqtgraph. We've largely held off from any kind of massive automated formatting efforts as it would create merge conflicts with existing PRs, which we have been chipping away at for a number of years now (we used to have almost 170 open PRs, currently under 30 half of which are draft PRs). The other reason is that pep8 doesn't quite line up with the Qt framework. For example, the Qt framework uses camelCase where as pep8 dictates snake_case. Due to our integration and dependence on the Qt framework, we adopt a lot of their naming convention.
When the PR count is effectively zero, we will be running black and pyupgrade on the entire codebase, and fixup any remaining flake8 errors; at which point we will also setup automated CI checks to check for those things (or perhaps use
pre-commit.ci to enforce those standards). We will need to do this carefully though, as running black on the entire codebase along would just entirely hose git-blame (there are ways to work around this). On the branch you linked to, I would be inclined to not merge that presently, as it would likely result in some merge conflicts for other PRs that change GraphicsScene.py, and any kind of git-blame check would point back to the commit you made as it stands now.
That said, we have done some formatting tasks for the entire codebase, not long ago we ran isort over the entire codebase, as well as pycln. Due to the limited scope of those runs, we were able to run those tasks w/o significant consequences to existing PRs.
Hope that answers your question and addresses your concern.