ValueError when running with newer Python / NumPy

[liamtoney]

Using Python 3.14 / NumPy 2.4.3, running example.py results in an error:

File /path/to/lts_array/lts_array/classes/lts_classes.py:727, in post_process(dimension_number, co_array_num, alpha, h, nits, tau, xij, coeffs, lts_vel, lts_baz, element_weights, sigma_tau, p, conf_int_vel, conf_int_baz)
    725 with np.errstate(invalid='raise'):
    726     try:
--> 727         sigma_tau[jj] = np.sqrt(tau[weights, jj, :].T @ residuals / (
    728             m_w - dimension_number))[0]
    729     except FloatingPointError:
    730         pass

ValueError: setting an array element with a sequence.

This appears to be due to a new way that NumPy is array vs. scalar assignment. Related: uafgeotools/array_processing#52 (comment) — we can remove the pin on Python version once this issue is fixed, since that pin is what prevents this issue