Fix UDOP dtype issue (#34180)

* Trigger UDOP tests

* Try forcing dtype in LayoutLMV3

* Do checks to see where uint8 is getting in

* Do checks to see where uint8 is getting in

* Found it!

* Add .astype(np.float32)

* Remove forced check, make fixup

* Checking where exactly the uint8 creeps in

* More checking on the uint8 issues

* Manually upcast in rescale()

* Remove UDOP trigger

src/transformers/image_transforms.py

@@ -123,11 +123,11 @@ def rescale(
     if not isinstance(image, np.ndarray):
         raise TypeError(f"Input image must be of type np.ndarray, got {type(image)}")
 
-    rescaled_image = image * scale
+    rescaled_image = image.astype(np.float64) * scale  # Numpy type promotion has changed, so always upcast first
     if data_format is not None:
         rescaled_image = to_channel_dimension_format(rescaled_image, data_format, input_data_format)
 
-    rescaled_image = rescaled_image.astype(dtype)
+    rescaled_image = rescaled_image.astype(dtype)  # Finally downcast to the desired dtype at the end
 
     return rescaled_image