Need a test validating the notion of a the state representation

In the `virtual_map_state.proto` file (in the CN repo), there is an implicit notion of how the state is exactly represented in the virtual map.

The BN cannot rely on this file directly, so it must use its own definition that is binary compatible with the CN's definition. Consequently, it has to depend on this implicit notion defined in `virtual_map_state.proto`.

We need to do two things:

- Make this notion explicit (i.e., properly document it).
- Create a test that ensures the definitions in `virtual_map_state.proto` comply with the set of rules that the BN relies on.

Here are the examples of such rules:
- if a key is a singleton, it's represented as a `SingletonType` encoded as `1`, and then it's follow by a specific number for the singleton keys.
- range of `1-50` is taken by regular keys
- range `126 - 10010` is taken by queue metadata
- if a key is not a singleton and not a queue, it's encoded with `oneof` range from 2 to 54, and then followed by the encoded key object
- queue states use `126 - 10010` range.