Data (column) context container #80

Create a context container(?) to manage the automation context on the frontend.

It could rely on a dataset just like the mappers do at this point.
Two options:

• Mapper dataset where both the source and target formats are known through the related fields datasets.
• Fields dataset only to provide context for the source data.

Edit 2026: Routines already support schema structures for both input and ouput. I think we should extend this to automations.

It should follow the route as a whole and provide autocomplete options to key params, conditionals and tags based on the current location in the flow.

Important dependencies:

• Field data types Define data field / column types #75
• Some form of processing of tasks to set the keys etc.