Facilitate adapter version

[lars-petter-hauge]

solves: #488

We don't know the version until the model has run (as it can change dynamically), hence we put it on the SimulationResult. Slight awkward that e.g. model_id and model_version are attached to two different pydantic objects (only mapped through Simulation_id), but that's perhaps something we need to live with.

[lars-petter-hauge]

not ment to be like this, was just to showcase that it works

[lars-petter-hauge]

Happy to see the tests are failing (with sensible error) without any migration setup!

self = <sqlalchemy.dialects.postgresql.psycopg2.PGDialect_psycopg2 object at 0x7f0584d2fec0>
cursor = <cursor object at 0x7f0585f5d990; closed: -1>
statement = 'INSERT INTO results (simulation_id, model_version, concentrations, panels, python_exception, error, id, created_at, u...ncentrations)s::JSON, %(panels)s::JSON, %(python_exception)s, %(error)s, %(id)s::UUID, %(created_at)s, %(updated_at)s)'
parameters = {'concentrations': '{}', 'created_at': datetime.datetime(2025, 11, 17, 15, 21, 3, 327652), 'error': "'tuple' object has no attribute 'model_version'", 'id': UUID('19828f44-4aae-4502-bdf9-550e76ca8cba'), ...}
context = <sqlalchemy.dialects.postgresql.psycopg2.PGExecutionContext_psycopg2 object at 0x7f0584d2e4e0>

    def do_execute(self, cursor, statement, parameters, context=None):
>       cursor.execute(statement, parameters)
E       psycopg2.errors.UndefinedColumn: column "model_version" of relation "results" does not exist
E       LINE 1: INSERT INTO results (simulation_id, model_version, concentra...