Avoid implicit array -> scalar conversion in tests (numpy 2.4)

[wishstudio]

Many tests fail with numpy 2.4

_____________________ TestCVodeReturn.test_jac_error_late ______________________
TypeError: only 0-dimensional arrays can be converted to Python scalars

The above exception was the direct cause of the following exception:

self = <odes.tests.test_user_return_vals_cvode.TestCVodeReturn testMethod=test_jac_error_late>

    def test_jac_error_late(self):
        solver = ode(self.solvername, complex_rhs, jacfn=jac_error_late, old_api=False)
>       soln = solver.solve([0, 1], [1])
               ^^^^^^^^^^^^^^^^^^^^^^^^^

src/scikits/odes/tests/test_user_return_vals_cvode.py:243:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
/nix/store/wyl71rbgix5j7lm559zm79raxwdif8fk-python3.13-scikits.odes-3.1.1/lib/python3.13/site-packages/scikits_odes/ode.py:177: in solve
    return self._integrator.solve(tspan, y0)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
src/scikits_odes_sundials/cvode.pyx:1819: in scikits_odes_sundials.cvode.CVODE.solve
    ???
src/scikits_odes_sundials/cvode.pyx:1857: in scikits_odes_sundials.cvode.CVODE._solve
    ???
src/scikits_odes_sundials/cvode.pyx:1383: in scikits_odes_sundials.cvode.CVODE._init_step
    ???
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

t = 0.0, y = array([1.]), ydot = array([1.])

    def complex_rhs(t, y, ydot):
>       ydot[0] = t - y
        ^^^^^^^
E       ValueError: setting an array element with a sequence.

src/scikits/odes/tests/test_user_return_vals_cvode.py:11: ValueError

This patch fixes the problem by using array[0] explicitly.