I think in most of the cases it is simple: Redshift exists for years, whereas databricks sql is pretty new.
So a lot of companies were already using redshift when dbrx sql popped up.
Once you have something in production, it is not an easy task to phase it out.  So when you decide to start using databricks sql, the redshift will still exist for a while.
F.e. we also have a Synapse database running while using db sql.  We want to get rid of it, but that takes some time because we have to migrate/adapt all queries running on that synapse

There might also be technical features that are possible and used in redshift, which are not (yet) available in db sql.