Uh, thanks for your 'Word in God's Ear', as the German language saying goes.
I think you might be referring to something that may have been published 30 or 40 years ago in the TABs (Technical Advisory Bulletins), which were a DH specific predecessor to Service Information Letters. Those documents are informal. They are not controlled, nor is a revision service (in the sense of how SBs and ADs are revised) provided.
The AFM, on the other hand, is formal, it is the only document related to the aircraft that is approved by both the regulatory authority of the State of manufacture and the regulatory authority of the State of registration.
The propeller governor has its own oil pump, therefore, there is no pressure loss when oil is routed to the propeller. There is a decrease in the volume of oil, but this is of no consequence during the start process, because all of the oil should be back in the tank at the beginning of engine start.
I assure you, there is no limitation on torque when in feather published in the AFM.
I disagree with your comment about not getting idle on the governing fuel flow prior to taking it out of feather. Many (but not all) of the PT6A-27 engines installed on the Twin Otter will idle on minimum flow at ISA pressure altitude until the propeller is taken out of feather. This is an unintended consequence of Py air being bled at the Nf governor when the propeller is feathered. But, not all engines have this characteristic, nor is this characteristic by design.
Michael