OK here is the chart I have been working on, green areas are possible with stock regulator , yellow I think I have found a replacement but going to have to change inductor also and red is external only. Now this is just for reference this is what worked for me, I did not go to far on the cooling aspect, no soldermask removal or anything other than the stock thermal pad. I am amazed that this chip can even handle this much power.
I'm curious as to how much of an effect cooling would have on stability at those high frequencies? My suspicion is that those voltages could be a bit lower starting @ around 20Mhz if better cooling is implemented. Not that it really matters because there is no practical way to power it. I do have one of the AOZ1036PI chips I had contemplated epoxying a thin copper sheet in the pad area so that it contacted the bottom LX pad then the LX area on the board. That chip is rated for 5A and would otherwise be a direct replacement, although the inductor may need to be swapped out as I'm not sure what current it is rated for. That may be out of the scope of some but it's an idea.
Anyway, Great work and thanks for posting up your results. I can confirm that your results match mine @ 14 16 and 18 as those are the one I've tried. I haven't had 20mhz working yet.