Fix indexing with bools

[brokkoli71]

fixes #1963

Changes:

• Mask indexing is available as indexing with square brackets: a[mask_selection]
• Orthogonal indexing with bool lists is available as indexing with square brackets: a[bool_selection_axis1, bool_selection_axis2]

Example
For an Array

data = np.arange(9, dtype="uint16").reshape(3, 3)
z = Array.create(
        StorePath(MemoryStore(mode="w")),
        shape=data.shape,
        chunk_shape=(data.shape),
        dtype=data.dtype,
    )
z[:] = data
mask = np.tile([True, False, True], (3, 1))

Each of the following failed before and works now

print(z[mask])
z[mask] = 42
print(z[[True, False, True], [True, False, True]])
z[[True, False, True], [True, False, True]] = 10

[jhamman]

A few quick comments.