Volker
Databricks Partner

I adjusted the schema in both bronze and silver, such that I do not need schema evolution. The problem is that the DataStreamReader does not pick up the schema changes in bronze.
I already figured out that it has something to do with providing also a 
starting_version since I tested it with and without starting_version and when I provided a starting version the stream did not pick up the schema changes and if omit the starting version then the stream does pick up the changes.
Now I'm a little confused since starting_version should only concern the stream when the query gets started and then should be ignored according to the documentation.