To follow Google-recommended best practices in Cloud Function development, what should you do before deploying changes?

Testing the function locally is critical before deploying changes, as it allows developers to validate the behavior and functionality of the code in an environment that closely mimics production. By running tests locally, you can catch any issues early in the development process, ensuring that your code behaves as expected without the overhead or risks associated with a live environment. This approach aligns with the principles of continuous integration and deployment, helping to maintain high code quality and reducing potential downtime or errors when the function is deployed to production.

Other options, such as erasing the original function or taking backups, may introduce additional risks or complications. Modifying logs does not contribute to the overall stability or functionality of the code being deployed. Therefore, local testing stands out as the most effective and recommended practice in ensuring the reliability of Cloud Functions before deployment.