Currently, if ESLint or any other (fixable) process fails, we just fail without giving a way to run the job again (via workflow_dispatch) and auto-fix the errors.

Ideally, the build-node-python takes another input to auto-fix errors which first runs the normal pipeline, and, in case of fixable errors and the flag being set (default false), runs the yarn lint:fix (for frontend) and make format (for backend) to fix the errors, runs, the lint again to ensure it's succeeding, and then (and only then) commits the changes with an appropriate commit message.