Gym-pbn

Primarily fix a couple of bugs, add some validation and a new evaluation function that does the evaluation style in the SDC paper.

**Full Changelog**: https://github.com/UoS-PLCCN/gym-PBN/compare/v1.1.0...v1.1.1

Migrate over to Gymnasium (v27)

- Added a lot of new environments (temporally abstract actions, options, target control, larger envs using Bittner data).
- Refactored the codebase significantly.
- Fixes and updates.

Initial version of the Probabilistic Boolean (Control) Network OpenAI Gym environment.

Contains full featureset, but no per-configured networks. Only base environments are included that can be parameterized with network data at environment instantiation time.