You are confusing a few different things.
First, CDo is a drag coefficient. To get a drag force, you need to multiply by V^2 from the dynamic pressure. So, a slight drop in CDo will still result in the behavior you expect.
Second, the chart you show is for a trimmed airplane -- lift equal to weight. So, the induced drag drops off because CL^2 is dropping off as the lift coefficient required for lift equal to weight decreases with increasing velocity.
Finally, the CDo change with velocity is really a Reynolds number effect. Your Reynolds number goes up, so cf goes down, so CDo goes down.
Rob