Agreed. You can at least do the single test rock and for the first time model a true kill step. Whether that's upscaled or not doesn't matter, it'll be firm knowledge one way or another. When plants dominate a scape, they reduce porosity of rocks, the ability to take in nutrients and for the tiny animals within to cast out their waste; even if a doser like fluconazole kills your algae, this clouding is cyano fuel
Or dino fuel, it's a feed for the next alternating generation to be determined. The cloudless, no organic stores of waste reef has to be attained physically eventually. Once plants are removed, the rocks are going to spend the next several weeks expressing pent up waste so this is one way external cleaning is a hidden boon... It's detailed removal and rinsing externally so you can avoid weeks of repeated export and water changes from the reclaimed tank
Anything claimed to kill that algae can be modeled ahead of time. If someone says X amnt of X added to the tank will kill the algae, that can be modeled in a test 5 gal bucket in dilution so that you can gauge death and growback cycles, if any, before you subject your nontargets in the tank to the experimental whatever
Test modeling might be really helpful even if peroxide isn't the selected tool. That stuff has such a weak design it's like Ulva algae and peroxide likely kills it overnite, that fast