Spike to sketch out what in data per trait versioning looks like

[SamCrooksFoundry]

What

Presuming an in-data per trait versioning schema sketch out a use case involving a DCC and a manager, communicating with mixed version traits.

Why

To ensure that the interfaces are reasonable and the workflows make sense.

Tease out any further decisions that have to be made before implementation.

AC

• Cover all methods that take trait ids/data as arguments.
• Create an executable sketch if it's feasible/sensible
• Both manager and host need consideration
• Check the implications of specification have no versions.
• Strategies for importing multiple versions of media-creation traits - in both python and cpp
• Update https://github.com/OpenAssetIO/OpenAssetIO-TraitGen/issues/80