Rapido are being cautious there. The particularly problematic output from Hammant & Morgan Duettes or Clippers is the 'half wave' selector which will give a coreless motor a good rattling at 50Hz, which is not best for their longevity.
Otherwise these controllers do what all 'resistance' type DC controllers do, and puts out rectified DC which is a 100Hz non-sinusoidal waveform. This isn't ideal for coreless motors, but a lot better than the 'halfwave' which is '10ms on, 10ms' off, which 'jerks' the low mass armature of a coreless motor around.
I have these very controllers, and the Duette is most useful for test running current OO models before decoder fitting, so that any mechanism shortcomings are detected and corrected. But there are two factors which make this safe. Firstly the resistance mats installed were the optional 'high' type, which significantly limits current; and then something I did long ago when iron core can motors - such as Mashima - became standard, was disconnect the 'half wave' output as redundant. I have test run all my RTR OO purchases these last 23 years on the Duette, and that includes coreless motored product from brands including Rapido, and had no consequent trouble: so they are not the 'spawn of satan', but do need to be used with thought. (The Clipper is in original condition and used when a 1960s or earlier OO model needs a good wallop to get it going, the old heavyweight iron core motors need the current this unit can deliver.)
Practically every other DC resistance type controller will be 'as bad' as the H&M product, though few if any will offer 'halfwave': but most won't have survived in usable condition! The H&M product was of pretty much 'bombproof' overall construction, and thus many still work...