Hello
I am using DeepSource to analyze a Python project & I’m running into a consistent false positive. It flags an unused import in a module where the import is dynamically accessed at runtime (specifically via globals() in a plugin loader). 
While the import appears unused in static analysis, it’s actually critical for the runtime execution.
I’ve tried using the # noqa and # skipcq directives, but I’m curious if there’s a better way to tell DeepSource this is a dynamic import context. 
I want to keep the benefits of clean linting without having to suppress real warnings across the board. I checked Issues - DeepSource Tableau Course guide related to this and found it quite informative.
Is there a recommended pattern or annotation to handle such dynamic use cases without compromising scan quality? 
Or is this something best handled through .deepsource.toml exclusions?
Thank you !! 