I seem to think that magnetic rapid start fluorescent ballasts are CWA ballasts because I notice that these ballasts produce the same flash patterns as CWA HID ballasts. Is this true?

I think so, as both requires a capacitor for proper lamp operation, as both uses the capacitor as their ballasting component, and both uses the same coil both for stepping up the voltage and regulating the current to the lamp.

The HPF ones are typically based off the CWA design. The NPF ones are like HX ballasts.

HX ballast is a reactor ballast so don't need a capacitor for proper lamp operation, so a NPF rapid-start ballast also don't require a capacitor if it is similar to HX.

The capacitor itself does not make it CWA.
CWA means there is the capacitor in series with the nonlinear inductive reactance formed by the saturating magnetic shunt between the primary and secondary.
I dont think most fluorescent ballasts aren't that complex, they use to be just a step up autotransformer with all winding together without any shunt whatsoever and then the series ballasting capacitor.

But making the arc output a CWA is indeed possible, but I dont think it is really the case, fluorescents, mainly in the RS circuits, dont need as accurate feed current as HID's do.

The HPF ones are typically based off the CWA design. The NPF ones are like HX ballasts.

I do have a single lamp preheat F40T12 HPF ballast that is an HX autotransformer ballast because the lamp’s flicker bands are similar to those found on HID lamps running on HX autotransformer ballasts.

Keep in mind I was referring to rapid start ballasts in reference to your original question. HPF rapid start ballasts generally have the capacitor wired in series with the lamp and behaves similar to CWA HID ballasts, lamp current affected by capacitor value, ballast will overheat if cap is shorter so it's reasonable to infer they have similar topologies.

I'm also kinda curious why you chose to revisit this topic a year later by replying with something (preheat ballasts) outside the scope of my original reply.

I seem to notice that the 2 lamp HPF series F40T12 rapid start ballasts do use a capacitor in series with both tubes if I am not mistaken, if that is the case, it is definitely a CWA ballast. I would also figure that many HO and VHO rapid start ballasts are also CWA ballasts too since they seem to run multiple lamps at the same current with varying arc voltages.

CWA is only when the circuit really utilizes the current regulation by the saturating element in a series LC. Because that increases the current crest factor, I doubt it would be used on the fluorescent (the thin filament with the sensitive oxide coating offers very little thermal inertia to handle the spiky current).
With the fluorescent ballasts the magnetic shunt will be operating way below its saturation, just to provide high impedance at higher frequencies. So it is lacking the regulation capability of the CWA, so to me it xan not be qualified that way.

The HO and VHO series are probably the closest thing to a CWA equivalent, although with a different regulation mechanism as Medved thoroughly explained.

I think using HO or VHO ballasts with HID lamps has been tried a few times over the years on L-G, mainly as an experiment.

Do the 1000mA F72T12 HO rapid start ballasts for street lantern tubes function as CWA ballasts too if it is only rated for just one particular tube wattage?

If RS ballasts are CWA, why is it that I have two that have a bad cap and run and behave exactly like NPF? They don't burn up or go over board on current like CWA will if you take cap away ...

The ballast is in either way made so it exhibits a significant inductance as its output impedance.
If you connect the lamp directly, like in a NPF ballast, this inductance is, what limits the current.
The CWA, as well as lead ballast has a capacitor in series. Because the capacitor has opposite phase shift, its reactance subtracts from the inductive one. If the capacitive is greater (normally the case), the resulting difference (Xl-Xc) is then the impedance responsible for the operating arc current limiting.
The difference between just a lead ballast vs CWA is, in CWA the inductive part operates slightly saturated, so the inductance responds on the cuirrent (larger current leads to lower inductance and vice versa). That way the conditions leading to larger currents (higher mains, lower arc voltage,...) cause the inductance to be more saturated, so lower inductance, so less is subtracted from the capacitor, so the resulting impedance gets bigger, preventing the current from increasing so much. The drawbacks are higher current distortion, so crest factpor, which wears electrodes faster, but as well way stronger magnetostriction of the saturated section, so louder ballast buzzing. The streetlight MVs have way heavier electrodes, which can handle the higher crest factor way better than the thin fluorescent filament and the noise is not of big problem on a streetlight than on an indoor fluorescent fixture.
In just a plain lead circuit the inductive part is far from saturated, so the inductance stays constant, so the real current remains proportional to the voltage drop across the ballasting LC.

So when the ballast is designed so the capacitive reactance (Xc) is about double the inductive (Xl) and the capacitor fails short circuit, the resulting ballasting would come just from the inductive part (Xl), which happens to be close to the original difference (if Xc = 2 * Xl , the (Xc - Xl) = Xl), so the arc current stays unchanged. Normally for safety reasons the typical ballasts use to be designed to have the current with just the inductance in the circuit not bigger than normal, to prevent severe overheating so prevent fire hazard for the case the capacitor fails short circuit. Normally the current uses to be a bit lower, so the lamp still light, but with less output.