Optapy

Upgraded to OptaPlanner 9.37.0.Final. Despite being a updated major version of OptaPlanner, the vast majority of OptaPy users are unaffected (unless you use your own Java jars with javax dependencies).

Fixes:

- Truncating floats to 32-bits
- Nullable variables
- Java Stub files are now included
- Inheritance in problem facts

Add support for Python 3.11 for `jpyinterpreter`, significantly improving Python 3.11 score calculation speeds.

Also optimize how calls are done in the interpreter, which can provide a modest improvement to score calculation speeds for some constraints.

Add the ConstraintVerifier API to allow testing of constraints. See https://www.optapy.org/docs/latest/constraint-streams/constraint-streams.html#constraintStreamsTesting for details.

This is the first release of `optapy` that includes `jpyinterpreter`, a module created to translate Python function bytecode to equivalent Java bytecode to massively increase performance by avoiding Foreign Function Interface (FFI) calls. You don't need to do anything in order to use it; it is on by default. Functions and classes that cannot be translated will be proxied to their CPython functions and types.

Bug fixes:

- Support ValueRange, CountableValueRange and entity value_range_provider
- Support arbitary PythonComparable as planning_id

- Bug Fix: SolverManager `solve` and `solveAndListen` will now show exceptions
- Bug Fix: A Uni, Bi, and Tri Constraint Stream can now join an UniConstraintStream

Dependency Upgrades:

JPype1 upgraded to 1.4.0, which fixes ambiguity errors when dealing with overloaded methods

New Features:

- get_class is not longer required when interacting with SolverConfig and Constraint Streams
- Pythonic version of Constraint Stream, Joiners, and ConstraintCollectors methods
- New decorators `custom_shadow_variable` and `variable_listener`, which can be used to create a custom shadow variable that changes when a geninue `planning_variable` (or another shadow variable) changes.