I had the same problem a while ago when Matplotlib got the new "Viridis" color scheme and I wanted to use this (and some more) in my application. Until now the format of the colormaps for the ImageView widget are pretty simple and, as Vincent wrote before, you can just add your own at the top of the GradientEditorItem.py.
The Matplotlib cmaps work a bit different and there're actually several modes how to define your colors. For example is "viridis" just a list of 256 RGB-tuples, while "cubehelix" is generated by a parameterizable function. So you have to convert each of these into Pyqtgraph's format. I started doing that, but after the first two I decided to just convert them all.
So ... in the attached file is the ImageView example from PyQtgraph, extended with a converter function that converts any Matplotlib colormap to PyQtgraph's format and a modified ImageView that "monkey patches" the new colormap into the context menu of the GradientEditorWidget. I hope you find it useful. :)
I planed to make a pull request with the converter added to colormap.py and an extended init-function for the GradientEditor that accepts additional colormaps. If somebody wants to include it before I find the time .... feel free. :)
Sebastian