“The preferred walking speed is the speed at which humans or animals choose to walk. Many people tend to walk at about 1.42 metres per second (5.1 km/h; 3.2 mph; 4.7 ft/s).”
“The results show teenagers walk at an average speed of 1.45 m/s, young adults walk at an average speed of 1.55 m/s, middle age pedestrians walk at a speed of 1.45 m/s, older pedestrians walk at speed of 1.09 m/s, and elderly or physically disabled pedestrians walk at a speed of 1.04 m/s.”
5km/hour is about 1.4m/s; the fastest of these speeds is 5.6 km/hour.
Initial kinetic energy is not the right physical quantity to look at. Most of that kinetic energy will remain in the car/bike/…, i.e. it doesn't tell you much about how much energy will get transferred to the victim – it merely gives you a bound from above.
No. The kinetic energy of the car/bike/… doesn't tell you anything because you don't know how much energy gets transferred to the victim until you have applied momentum conservation to the elastic/inelastic problem. So, the right approach would be (in this order): Calculate momenta, calculate how much momentum gets transferred to the victim via momentum conservation, deduce the resulting change (increase) in kinetic energy of the victim. This kinetic energy will be converted into heat (= damage/injuries) in one way or another, so it's the relevant physical quantity for our considerations.
Finally, in case of an inelastic problem (likely with cars, not so likely with bikes or people), you also need to consider the energy loss during momentum transfer. Once again, this energy will become heat (= do damage), so it adds to the aforementioned increase in kinetic energy when we're interested in how much damage will be done.
Walking: 5 km/h * 70kg => 875 (although this is a very slow estimate for a walker) (please ignore the non-canonical units)
e-bike: 20 km/h * 90 kg => 18,000 (I think your estimate for ebike mass might be on the low side at 20kg, but whatever)
car: 50 km/h * 2000kg => 2,500,000 (and that's a fairly low speed for cars! Drunk drivers often drive faster than is wise.)
Everything else is a rounding error compared to the energy of a car.