Make nice ranges for IDs we use for system serializers

Make sure Ids we use for serializers make sense and are grouped up nicely and are possible to evolve/add into each group in the future. I.e. don't just 1,2,3,4,5 in there.

The issue names no files, serializer definitions, or tests. Start by locating the system serializer ID declarations and their existing uses, then determine how IDs are currently assigned and where compatibility is verified; done means the IDs are grouped into sensible, extensible ranges without breaking existing serialization.

Written by the indexing model from the issue text.