[Task] SQL generator: Handling Collections of Entities

[atextor]

Is your task related to a problem? Please describe.
Currently, for a Property that has a Collection/List/Set/SortedSet Characteristic, the Databricks SQL generator will create one column with an ARRAY<> datatype, with a STRUCT for the nested Entity type. This makes querying difficult and the approach is not applicable to other SQL dialects.
Instead, such collections of Entities should be flattened into columns for the Entity's type, with additional discriminator columns that allow retrieving the original dataset. Additionally,
for Aspects with one Property called items (a "collection Aspect"), flattening to tables should happen for the Entity which is the type of the collection, not for the Aspect itself. As an example, for AspectWithErrorCollection.ttl, the resulting table should have columns for errorNo, errorText and startTimestamp instead of an items column.

Describe the solution you'd like
Adjust the SQL Databricks generator to flatten collections of entities into columns. Handle a top-level items Property separately if necessary. Make names of additional discriminator columns configurable.