Redo constraints (since columns can only take one constraint at create time)

[benc-db]

Description

When I went to apply the constraint logic from the previous PR, I discovered that Databricks does not like receiving multiple constraints for a single column in the table create syntax; however, you can add arbitrarily many table level constraints, so in this PR I aim to do the following:

• if a column has a not_null constraint, directly set not_null on the column
• any other column constraint gets converted to a TypedConstraint, a subclass of ModelLevelConstraint that enforces the requirements of the particular constraint, and holds the rendering logic. The conversion required setting 'columns=[name]'. Because python makes operator overloading a pain in the butt, I'm going back to OOP for the render logic.
• If a model level constraint is not_null, convert to setting not_null on each affected column, since Databricks does not support not_null constraint definition as a table constraint
• any other model constraint gets converted to a TypedConstraint

Checklist

• I have run this code in development and it appears to resolve the stated issue
• This PR includes tests, or tests are not required/relevant for this PR
• I have updated the CHANGELOG.md and added information about my change to the "dbt-databricks next" section.

[alexguo-db]

LGTM

[alexguo-db]

Is this used anywhere?

[benc-db]

Only in my branch where I found out it doesn't really work. I tried to isolate model and column constraint parsing/processing before, but when I tried to use the constraint code on my local branch, I discovered that Databricks does not like multiple column constraints on a single column; also dbt supports specifying a bunch of null constraints at the model level. Hence why I have to do this less elegant, 'process them all at the same time' approach :/.