Thank you for your reply! I'm not sure if audit logs help in this case. I meant tracking changes more in the sense of code versioning. I found this answer https://community.databricks.com/t5/data-engineering/how-to-deploy-azure-databricks-sql-table-views-... which looks promising but I'm curious if there are better solutions. 
Ideally there would be some solution similar to terraform that detects and applies changes to a table and can, unlike terraform, run "ALTER TABLE ..." statements to adapt the schema without dropping the original table.