[sumitsaraswat362]

Fixes #750.

Description

In the Numba Object backend, several coordinate conversions and helper methods available on the Python Vector objects were missing corresponding @numba.extending.overload_method registrations, causing Numba JIT typing failures.

This PR implements the missing overload dispatchers in src/vector/backends/_numba_object.py:

1. Numbered dimension conversions (to_2D, to_3D, to_4D):
   • Registered @numba.extending.overload_method for to_2D, to_3D, and to_4D directly across VectorObject2DType, VectorObject3DType, and VectorObject4DType, mapping to the corresponding to_Vector*D handlers.
2. Momentum coordinate conversions:
   • In add_coordinate_change, registered the corresponding momentum method name alongside the generic method name (e.g. to_pxpy, to_ptphi, to_pxpypz, to_pxpypzenergy, to_ptphietamass, etc.), reusing the exact coordinate transformation closure.
3. like(other) dispatch:
   • Added @numba.extending.overload_method for like on 2D, 3D, and 4D vector types with typing-time inspection on other:
     • VectorObject2DType -> dispatches to v.to_Vector2D()
     • VectorObject3DType -> dispatches to v.to_Vector3D()
     • VectorObject4DType -> dispatches to v.to_Vector4D()
     • Non-vector types cleanly raise numba.core.errors.TypingError at typing time.

Testing

• Added comprehensive unit tests in tests/backends/test_numba_object.py covering:
  • to_2D, to_3D, to_4D dimension changes inside JIT functions.
  • Coordinate conversions to Cartesian, Polar, Cylindrical, and Momentum representations across 2D, 3D, and 4D vectors inside JIT functions.
  • like() conversions matching 2D, 3D, and 4D vector targets.
  • Rejection of non-vector targets raising numba.core.errors.TypingError.
• All 48 tests pass locally. Linted cleanly with ruff.