Fixes mypy failure with latest mypy 1.16.0

[woodsp-ibm]

Summary

Fixes for latest mypy errors in CI

Details and comments

Was erroring as follows

qiskit_machine_learning/neural_networks/effective_dimension.py:72: error: Unused "type: ignore" comment  [unused-ignore]
qiskit_machine_learning/neural_networks/effective_dimension.py:74: error: Unused "type: ignore" comment  [unused-ignore]
qiskit_machine_learning/neural_networks/effective_dimension.py:86: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[floating[_64Bit]]]", variable has type "None")  [assignment]
qiskit_machine_learning/neural_networks/effective_dimension.py:98: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[Any]]", variable has type "None")  [assignment]
qiskit_machine_learning/neural_networks/effective_dimension.py:112: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[floating[_64Bit]]]", variable has type "None")  [assignment]
qiskit_machine_learning/neural_networks/effective_dimension.py:124: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[Any]]", variable has type "None")  [assignment]
qiskit_machine_learning/neural_networks/effective_dimension.py:150: error: Need type annotation for "param_set"  [var-annotated]
qiskit_machine_learning/neural_networks/effective_dimension.py:334: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[floating[_64Bit]]]", variable has type "None")  [assignment]
qiskit_machine_learning/neural_networks/effective_dimension.py:349: error: Incompatible types in assignment (expression has type "ndarray[Any, dtype[Any]]", variable has type "None")  [assignment]
qiskit_machine_learning/kernels/algorithms/quantum_kernel_trainer.py:221: error: Incompatible types in assignment (expression has type "Union[Any, float]", variable has type "int")  [assignment]
Found 10 errors in 2 files (checked 188 source files)

The last one above, in quantum_kernel_trainer I fixed it by correcting the type of a field it was accessing when creating the result where the field in question is part of the results parent class VariationalResult in variational_algorithm. The optimal_value had a type mismatch between the getter and the setter.

[woodsp-ibm]

It seems to pass ok with Python 3.9, which is what I have locally and what I tested with, but in other versions of Python there is an additional failure

qiskit_machine_learning/gradients/lin_comb/lin_comb_estimator_gradient.py:236: error: Incompatible types in assignment (expression has type "ndarray[tuple[int, ...], dtype[Any]]", variable has type "ndarray[tuple[int], dtype[Any]]")  [assignment]
Found 1 error in 1 file (checked 186 source files)

that needs correcting.

Update: for this I simply added a type ignore.
Well it seems that this is fine for > 3.9 but under 3.9 it ends up with an unused ignore error - sigh!

qiskit_machine_learning/gradients/lin_comb/lin_comb_estimator_gradient.py:236: error: Unused "type: ignore" comment [unused-ignore]

Finally I updated the ignore to cover both cases so it ignores the fact that its unused under 3.9 as well as covers the assignment type aspect for > 3.9.

[coveralls]

Pull Request Test Coverage Report for Build 15329190616

Details

• 7 of 7 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 90.821%

---

Totals
Change from base Build 15279272116:	0.0%
Covered Lines:	4492
Relevant Lines:	4946

---

💛 - Coveralls