port the numpy.ma typing improvements from NumPy

[jorenham]

The following can a bit of a package-deal, affecting ma[.MaskedArray].{[arg]min,[arg]max,ptp}

• TYP: Type masked array shape, dtype, __int__, and __float__ numpy#28591 - 🚚 port type masked array shape, dtype, __int__, and __float__ #472
• TYP: Type numpy.ma.min numpy#28593 - 🚚 port numpy.ma.min #474
• TYP: Type ma.max and ma.ptp numpy#28612
• TYP: Type ma.MaskedArray.min numpy#28616
• TYP: Type ma.MaskedArray.max and ma.MaskedArray.ptp numpy#28623
• TYP: Fix overload for ma.MaskedArray.{min,max,ptp} and ma.{min,max,ptp} for positional axis and out numpy#28637
• TYP: Type MaskedArray.{argmin, argmax} and np.ma.{argmin, argmax} numpy#28638 - 🚚 port ma.arg{min,max} and MaskedArray.arg{min,max} #468
• TYP: Type MaskedArray.sort numpy#28664
• TYP: Type MaskedArray.partition and MaskedArray.argpartition numpy#28677
• TYP: Type MaskedArray.take and np.ma.take numpy#28680
• TYP: Type MaskedArray.__{ge,gt,le,lt}__ numpy#28689
• TYP: Type np.ma.{is_masked,ndim,size,ids,iscontiguous} numpy#28715
• TYP: Type MaskedArray.count and np.ma.count numpy#28735
• TYP: Type MaskedArray.filled and np.ma.filled numpy#28742
• TYP: Type MaskedArray.put, np.ma.put, np.ma.putmask numpy#28745
• TYP: Type MaskedArray.compressed and np.ma.compressed numpy#28766
• TYP: Type np.ma.allclose and np.ma.allequal numpy#28793
• TYP: Type np.ma.squeeze and np.ma.{mask_rows, mask_cols, mask_rowcols} numpy#28794
• TYP: Fix type annotations for np.ma.nomask and np.ma.MaskType numpy#28831
• TYP: Type np.ma.getmask numpy#28832
• TYP: Type np.ma.is_mask numpy#28833
• TYP: Type np.ma.nonzero and MaskedArray.nonzero numpy#28845
• TYP: Use _Array1D alias in numpy.ma.core.pyi numpy#28847
• TYP: Type np.ma.ravel and MaskedArray.ravel numpy#28848
• TYP: Type MaskedArray.repeat, improve overloads for NDArray.repeat, generic.repeat, and np.repeat numpy#28849
• TYP: Type MaskedArray.swapaxes numpy#28850

[jorenham]

Note that all of these have been authored by @MarcoGorelli