Refactor: misc best practices

[sbrugman]

Split each refactor per commit for easier review:

• Removed unused parentheses
• Added missing super().init() calls
• Export visitors for more concise imports
• Removed dead code (unused assignments)
• Instead of suppressing type checking, provided type hints to mypy via assert where possible
• Used unpacking instead of concatenation for tuples
• Simplified the control flow where possible (reducing nested blocks etc.)
• args.path is already a required argument and validated by argparse, removed not args.path which is never reached

I'll rebase/group the commits and fix test when the other PRs are merged. I've got more functional changes lined up, but will hold off until you've had time to review/merge the outstanding PRs to avoid conflicts.

[kit1980]

Let's not reformat test fixtures. They are not always even correct Python.

[sbrugman]

@kit1980 I'll remove the formatting of test cases from: #69 and update this PR accordingly. Indeed the test fixtures are valid Python syntax, but not always functioning (e.g. missing required arguments)

[sbrugman]

None of the proposed changes in this PR touch the test suite. I've separated them from the other PRs, so can be reviewed independently. All fairly minor if you see the changes in isolation (per commit).

[kit1980]

Thanks!

[sbrugman]

@kit1980 could we please enable "Rebase and merge" in the settings? Squash is a good default, but in this case the commits are atomic, and it is helpful to keep them in case one type of changes need to be reverted/blamed (e.g. mypy). Both keep linear history.

[kit1980]

could we please enable "Rebase and merge" in the settings

done

[sbrugman]

Thanks!