Distributive members are swapped compared to Haskell/Purescript

[lemastero]

Implementation of Distributive looks like this:

trait Distributive[F[_]] extends Functor[F] { self =>
  def distribute[G[_]:Functor,A,B](fa: G[A])(f: A => F[B]): F[G[B]]
  def cosequence[G[_]:Functor,A](fa: G[F[A]]): F[G[A]]
}

But in Haskell and Purescript this looks differently:

collect :: Functor f => (a -> g b) -> f a -> g (f b)
distribute :: Functor f => f (g a) -> g (f a)

So probably in Scala implementation should be more like this:

trait Distributive[F[_]] extends Functor[F] {
  def collect[G[_]:Functor,A,B](fa: G[A])(f: A => F[B]): F[G[B]]
  def distribute[G[_]:Functor,A](fa: G[F[A]]): F[G[A]]
}

well, if we want to have co-name then maybe:

trait Distributive[F[_]] extends Functor[F] {
  def cotraverse[G[_]:Functor,A,B](fa: G[A])(f: A => F[B]): F[G[B]]
  def distribute[G[_]:Functor,A](fa: G[F[A]]): F[G[A]]
}

Maybe I am missing something :)

[lemastero]

Cats Distributive also has this feature/bug. Relevant PR: typelevel/cats#2046