A couple of things to keep in mind about this.
First, the database tag name attribute is a Torque id and not actually tied to any physical DB. For example, I have multiple Torque schema files for tables that all exist in the same physical DB but are used by subsections of my code.
Second, IMHO the schema files generated via JDBC should be considered a starting point for creating production ready files. There is just too many options that JDBC can know nothing about such as idmethod, java naming methods, inheritance, and the like. So the generated files will always need some hand tweaking to be right.
FWIW, I don't think there is a "right" answer to what goes in the generated name attribute. I could make the argument that as an MSSQL user who can only create schema files by specifying the owner of the tables that the dbuser is the correct value to use. Oracle users might claim they want the Oracle schema being mapped used. Who's right?
Maybe the best solution is to just used a static value like "jdbc-generated"? Perhaps this should be a new property in the default.properties that can be overridden in the build.properties?