Simplify X-basis Hamming weight state initialization

(Anthropic) Claude has greatly simplified the X-basis Hamming weight state initialization:

def init_thresholds(n_qubits, theta=0.0):
    n_bias = n_qubits + 1
    p = (1.0 - np.sin(theta)) / 2.0
    return np.array([math.comb(n_qubits, k) * (p ** k * (1.0 - p) ** (n_qubits - k)) for k in range(n_bias)])

The theta parameter is 0.0 by default, but OTOC(2) sampling uses it for initialization.

Full Changelog: v9.16.9...v9.16.10

sha1sum results:
28834a9cc1f8c1d868ebd31464dbaed3b2a44fbd pyqrackising-9.16.10-py3-none-macosx_14_0_arm64.whl
36a7cb0ae2e7d976dc14b0b11b2a180750ebf4c9 pyqrackising-9.16.10-py3-none-macosx_15_0_arm64.whl
4a70142a26aa90c4a28335852a62741633ffa2cd pyqrackising-9.16.10-py3-none-manylinux_2_35_x86_64.whl
7d41912e1db263dab1b539afeee9688d66765355 pyqrackising-9.16.10-py3-none-manylinux_2_39_x86_64.whl
333486c3271d5a12aaadf640e4187af600d0757e pyqrackising-9.16.10-py3-none-win_amd64.whl
3ec50c0fbbf100154ada9f7bfae2f3a86700bf53 pyqrackising-9.16.10.tar.gz