I have a general question regarding how the table in Databricks changes when we change the underlying data file in S3. For example, if I convert multiple folders that are defined by dates in a single folder into parquet by using the following code: C...