I dont think that this is possible. You could try aggregating tx/rx of each interface in steps.
datasource1 = Int1 tx
datasource2 = Int1 rx
datasource3 = Int1+Int2 tx
datasource4 = Int1+Int2 rx
datasource5 = Int1+Int2+Int3 tx
datasource6 = Int1+Int2+Int3 rx
and so on creating as many datasources as you need to cater for all your link tx/rx sources.
The code for the above sources could make use of this code Selfmade function: Get Aggregation-Link-Speed - help needed
But this would still be overlayed and would probably not give you the exact result that you are looking for. But if you just need a visual display it could look ok.. tbh, this is just theoretical, I have not tried it out yet, but might well give it a go when I get some spare time. I would be interested to see what it would look like.
Lebowski, do you see this method possibly working?