Bug#1029693: python-bumps: tests fail with scipy 1.10

[Drew Parsons]

Package: python-bumps
Version: 0.9.0-2
Severity: normal

scipy 1.10 is now available in experimental.
bumps fails debci tests using it.

We are considering uploading scipy 1.10 to unstable in order to
included it in the forthcoming stable release.  If we proceed with
that, then this bug will become Severity: serious.

The errors are AttributeError: 'multivariate_normal_frozen' object has
no attribute 'cov'
A sample from the failing test log is

_____________________________________ test _____________________________________

    def test():
        """check entropy estimates from known distributions"""
        # entropy test is optional: don't test if sklearn is not installed
        try:
            import sklearn
        except ImportError:
            return
    
        # Smoke test - do all the methods run in 1-D and 10-D?
        _check_smoke(stats.norm(10, 8))
        if hasattr(stats, 'multivariate_normal'):
            _check_smoke(stats.multivariate_normal(cov=np.diag([1]*10)))
    
        D = stats.norm(10, 8)
        _check_entropy("N[100,8]", D, N=2000)
        _check_entropy("N[100,8]", D, N=12000)
        if hasattr(stats, 'multivariate_normal'):
            D = stats.multivariate_normal(cov=np.diag([1, 12**2, 0.2**2]))
            _check_entropy("MVN[1,12,0.2]", D)
            D = stats.multivariate_normal(cov=np.diag([1]*10))
            _check_entropy("MVN[1]*10", D, N=10000)
            # Make sure zero-width dimensions return H = -inf
            D = MVNSingular(cov=np.diag([1, 1, 0]))
>           _check_entropy("MVN[1,1,0]", D, N=10000)

/usr/lib/python3/dist-packages/bumps/dream/entropy.py:1074: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
/usr/lib/python3/dist-packages/bumps/dream/entropy.py:1010: in _check_entropy
    H = D.entropy()/LN2
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <bumps.dream.entropy.MVNSingular object at 0x7fda5ba23d60>, N = 10000

    def entropy(self, N=10000):
        with np.errstate(divide='ignore'):
>           return 0.5*log(np.linalg.det((2*pi*np.e)*self.dist.cov))
E           AttributeError: 'multivariate_normal_frozen' object has no attribute 'cov'

/usr/lib/python3/dist-packages/bumps/dream/entropy.py:863: AttributeError