I'm going to answer it a bit differently, and perhaps not in the way that you want:
I don't think the question is a valid question to begin with. I know this may seem as if I'm calling you ignorant---nothing is further from the truth!---but perhaps one quick example can show that the question is not as easy to answer as you might think:
The following excerpt is from Bach's C-minor Fugue from Book I of the Well-Tempered Clavier:
The first two measures certainly seem to suggest C harmonic minor, right? So when we're compiling data for this hypothetical corpus study to determine which form of the minor scale is most common, this gets one check for C harmonic minor.
But then comes m. 3. The down-stem pitches are the collection of C melodic minor, but they're the ascending form of C melodic minor...and this scale is descending!
So where does this fit in our corpus? Do we check the melodic minor box, even though it doesn't actually follow the "rules" of melodic minor?
As we continue looking at more music, we'll keep seeing examples like this that don't fall neatly into one of the three forms of minor; because they don't map onto real works neatly, the hypothetical study in question cannot be done.
Think of "minor" as a compositional environment. Natural, harmonic, and melodic alterations are just that: little nuances that give flavor, but nonetheless do not alter the basic "minor" status.