It is possible for it to be a new mutation but in reality it's far, far more likely that a silky feathered bird was part of the silky feathered cochin's ancestry.
Crossing the two breeds- silky and cochin is not rare and has been known to be done on purpose to introduce something such as color or some part of the type.
And then it's very easy to get rid of crest, extra toe and the dark skin after the first cross.. Silky feathering is recessive so that one is harder 'to get rid of' as one cannot tell just by looking at the bird if it has silky feathering or not. That's how it can pass on for generations "undetected".
One example if a breeder uses one or very few roosters in their program, uses many hens and inbreeds/linebreeds. If the breeder chooses roosters that by coincidence does not have the silky gene.. he will never throw silky chicks, even if some of the hens have it.. even by line breeding daughters and grand daughters to him, no silky chicks even though some of the daughters are getting it from their carrier mother. This is one way how it can seem to be "no silky birds for many generations".
Especially so if breeder has a flock of multiple roosters and hens(either at same time or over time- such as getting new rooster for "new blood") and doesn't/refuses to inbreed in any form. This pretty much prevents any chance of a male and female carrying a silky feather gene from "meeting up" and throwing 25% sillky feathered chicks. So... many generations of this kind of set up has extremely low chance of showing any silky feathered birds even though the gene is in there.
Important to understand the "first cross" can either be recent or many generations ago.