If that was going to be a problem, I doubt GM would have attached the coils to a hot valve cover. The valve cover is also a heat sink, even if it's a hot one around 220-250 degrees. You don't need air to transfer heat, there is also conduction and radiation heat transfer. If the coil temperature is greater than the valve cover temperature, you will get heat flow to the valve cover. Spreading the heat transfer across 8 coils also helps. Also, what's going to fail? The coil would have to get hot enough to melt or somehow degrade the insulation between the windings. I'll bet a coil can withstand a few hundred degrees before it fails. I wouldn't worry about it.
The coil doesn't fire the injectors, but you knew that. Also, I doubt the power dissipated is actually as high as 30W per coil. I just used the fuse size as kind of an upper limit for the sake of discussion. It's probably quite a bit less in practice.PS. I don't really believe the 30 watts needed per coil will be expended as heat from the coil, since the majority of the energy is used for the injector to 'squirt fuel', but there will be 'some' residual heat from each pulse of the coil.. I haven't been able to ascertain an precise number for that as yet, but still believe without an 'air source' heat buildup can become a problem.