Ability to load configuration for deployment

[sliverc]

It is not clear yet how this is best implemented.

Issue is that user may update configuration in the user interface but on the other hand does a developer might need to implement a specific workflow on a development machine.

Those two should not overwrite but expand each other.

Options:

• using graphql files which are loaded
• using loaddata/dumpdata json files

Both have the issue that user data might be overwritten by dev data (e.g. a change of a name). Hence it might be an option to split structure and data from each other.

[sliverc]

An automatic possibility would be a CRDT algorithm like https://arxiv.org/pdf/1608.03960.pdf

But most likely this will cause errors as well as thing might be merged unpredictably.

Potentially we will need to think about marking certain part of configuration as dev managed and some as freely configurable.