[PyROOT] Avoid using deprecated numpy._float

The `np.float_` was removed in the NumPy 2.0 release. One should use
`np.float64` instead, which is already in the same dictionary.

Author: guitargeek

bindings/pyroot/pythonizations/python/ROOT/_pythonization/_rdf_conversion_maps.py

@@ -60,7 +60,6 @@
     numpy.int16 : 'Int_t',
     numpy.int32: 'Int_t',
     numpy.int64 : 'Long_t',
-    numpy.float_ : 'Float_t',
     numpy.float32 : 'Float_t',
     numpy.float64 : 'Double_t',
 }