Pull request type

• Code changes (bugfix, features)
• Code maintenance (refactoring, formatting, tests)

Checklist

• Docs have been reviewed and added / updated
• Lint (black rocketpy/ tests/) has passed locally
• All tests (pytest tests -m slow --runslow) have passed locally

Current behavior

1. The input parameters of the Funciton class are all checked in a very disorganized way. This would lead to redundant checks and overall slowing down of the class
2. all interpolations used in get_value_opt were using np.searchsorted, which was slow

New behavior

1. Validation has been reworked as to follow the order: source -> interpolation and extrapolation -> inputs, outputs and title. The get_value_opt function is now modularized and is defined according to the current interpolation and extrapolation
2. bisect_left is now used instead of np.searchsorted. This cuts the execution time by half in a u_dot_generalized call.

The Function class should behave exactly the same as before.

Breaking change

• Yes
• No

Additional information

• There is an attribute called __img_dim__ that was never used and was always equal to 1. I changed it so its defined at the __init__ for backward compatibility purposes, but it is still never used
• Further optimizations can be done in __mul__, __sub__, __truediv__ and other operations by avoiding going through all the validation checks when re-initializing the Functions