https://www.pinterest.com/pin/168533211025982771/https://www.pinterest.com/pin/556827941407233444/Technically, you want 'curves' not angles, at least until you've transitioned from the straight plank sides into the critical slope angle. Just as you've drawn for the side elevation.
The sides need to be exactly like the top, when viewed from above.
Only you can tell how much the diffuser angle will have to be compromised.
18% body length diffusers are never more than around 4-degrees upsweep.
The low drag diffusers are even 'slower' in angle if 'straight-flat'. They can withstand higher angles if gently and progressively curved, as in the Volkswagen 2000 concept of 1980. Much like your drawing.
I realize that is complicates fabrication by orders of magnitude, but in order to protect the boundary layer, there cannot be any 'kinks' or pressure 'spikes'.
The Evalia is reported at Cd 0.31. It's frontal area is going to be in the neighborhood of 2.9 meters-squared.
It's rated at 165- km highway in 'warm' weather.
391- Wh/mile ( 242.4 Wh/ km )
If we knew the exact speed at which your Evalia registered 242 Wh/km, that would be a good test speed.
------------------------------------------------------------------------------------
Dirty math suggests:
* Cd 0.31 OEM
* Cd 0.304 with rear skirts
* Cd 0.269 with smooth rear belly section and 'slow' diffuser
* Cd 0.17 with boat tail.
* a 45% drag reduction
* maybe 195-km range
You'd have to have the more complex geometry.
Here's how Union-Pacific Railroad handled the more complex surfaces in 1934. See link at top