There is a lot of fuzzy understanding around how all the canids are related. New genetic techniques have shown us that they have been interbreeding periodically pretty much forever; sometimes naturally and sometimes via selective crosses by humans. For example, the canine gene for black fur was selected by humans after dogs were domesticated. It did not exist at any prevalence in wolves (who share a common ancestor with dogs but we now know are not direct ancestors of them) until those domestics bred with wolves (we're talking ancient history here, not long after domestication began). Dogs, wolves and coyotes are all closely enough related to breed and produce fertile offspring. Those offspring will then usually back cross with one of the parent types, over time diluting the cross but maintaining some new genes that apparently convey some sort of selective advantage in the population. Coywolves are unusual but probably not as rare as people think, and are likely a result of an animal not having access to a mate of the same species for some reason- maybe a social outcast, maybe limited population size. Either way, these crosses would not be considered a separate species. Just hybrids. If those hybrids turn out to be selectively superior than the parent stock and outcompete them, then you may see speciation occur.
And yes, polar bear/ griz crosses are becoming more common as ice pack shrinks and polars are restricted to land during the breeding season instead of being out to sea.