MAINT: move to macos-15-intel images, and update some dependencies

[rgommers]

See commit messages for more details.

[rgommers]

Hmm one failure, same as numpy/numpy#29391. The scipy-openblas bump to 0.3.30.2 isn't actually robust, so need to drop that change. I'll document that in the requirements file instead, otherwise we'll rediscover that again next time someone tries.

EDIT: or not, unclear if 0.3.30.1 has the same problem, since apparently it has existed for a while, so this may just be the cause of a flaky test. Failure in one of the Windows jobs:

EDIT 2: it's not even on this PR, it's on the cibuildwheel bump merge to main that was just merged. So really just a flaky test.

  ___________________ TestRegression.test_openblas_threading ____________________
  [gw3] win32 -- Python 3.13.8 C:\Users\runneradmin\AppData\Local\Temp\cibw-run-r0nwspyh\cp313-win_amd64\venv-test\Scripts\python.exe
  
  self = <numpy.linalg.tests.test_regression.TestRegression object at 0x00000189A476FAD0>
  
      def test_openblas_threading(self):
          # gh-27036
          # Test whether matrix multiplication involving a large matrix always
          # gives the same (correct) answer
          x = np.arange(500000, dtype=np.float64)
          src = np.vstack((x, -10 * x)).T
          matrix = np.array([[0, 1], [1, 0]])
          expected = np.vstack((-10 * x, x)).T  # src @ matrix
          for i in range(200):
              result = src @ matrix
              mismatches = (~np.isclose(result, expected)).sum()
              if mismatches != 0:
  >               assert False, ("unexpected result from matmul, "
                      "probably due to OpenBLAS threading issues")
  E               AssertionError: unexpected result from matmul, probably due to OpenBLAS threading issues
  E               assert False
  
  expected   = array([[-0.00000e+00,  0.00000e+00],
         [-1.00000e+01,  1.00000e+00],
         [-2.00000e+01,  2.00000e+00],
         ...99997e+06,  4.99997e+05],
         [-4.99998e+06,  4.99998e+05],
         [-4.99999e+06,  4.99999e+05]], shape=(500000, 2))
  i          = 0
  matrix     = array([[0, 1],
         [1, 0]])
  mismatches = np.int64(21849)
  result     = array([[ 0.00000e+00,  0.00000e+00],
         [-1.00000e+01,  1.00000e+00],
         [-2.00000e+01,  2.00000e+00],
         ...99997e+06,  4.99997e+05],
         [-4.99998e+06,  4.99998e+05],
         [-4.99999e+06,  4.99999e+05]], shape=(500000, 2))
  self       = <numpy.linalg.tests.test_regression.TestRegression object at 0x00000189A476FAD0>
  src        = array([[ 0.00000e+00, -0.00000e+00],
         [ 1.00000e+00, -1.00000e+01],
         [ 2.00000e+00, -2.00000e+01],
         ...99997e+05, -4.99997e+06],
         [ 4.99998e+05, -4.99998e+06],
         [ 4.99999e+05, -4.99999e+06]], shape=(500000, 2))
  x          = array([0.00000e+00, 1.00000e+00, 2.00000e+00, ..., 4.99997e+05,
         4.99998e+05, 4.99999e+05], shape=(500000,))

[charris]

That error isn't unusual these days. Need to ping Apple again.

EDIT: It is common enough that we might want to ignore that failure in the test.

[charris]

Thanks Ralf.