Avoid duplicate findings from multiple sarif files

Overview

Avoid duplicating preexisting findings when processing multiple (potentially duplicate) SARIF files for a given tool

The basic solution is to use IndexedSet to enforce uniqueness when combining ResultSets
- IndexedSet preserves ordering 👍
In order to make this possible, the Result type needs to be hashable. This required the following changes
- Using frozen datatypes
- Using frozen pydantic classes
- Replacing all instances of list (not hashable) in these types with tuple (hashable)
The type changes introduce a few downstream breaking API changes, but they should be easy to absorb