using odbc as Hubert mentioned is the easiest way.

Besides Databricks and Databricks SQL, there are also other options like Azure Synapse Serverless, Presto etc.

They all serve data as tables.

Python also has a parquet reader using pyarrow f.e.