Ok so first problem, those horns. Many people here have mentioned horns are a patterned herbivorous trait, but they seem to forget that many reptiles have at least ornamental horns. This is none the less not a great problem, as a horned, herbivorous ancestor developing an obligate carnivorous lifestyle isn’t far fetched (I would say this is the most likely path).
The real problem comes in two parts: living examples and plain physics.
1: I cannot count a single vertebrate that impales prey. Shrikes don’t count, as the impaling implement is not on their body. Even insects that impale (true bugs) do so to inject venom and suck up nutrients - your bicorn gains neither benefit. An impaling implement needs a grasping or dragging implement to un-impale the prey item - can’t do that with hooves. There isn’t much benefit to the fitness of an animal to evolve an impaling horn, especially above and out of reach of their mouth. Side note: incase you mention them boar and other tusked animals, as well as likely ceratopsians and carnotaurus, do not impale but gore - usually in defence or territorial dispute.
2: Inefficiency. Simply put, when puncturing something, two heads are worse than one. Cutting or puncturing effectiveness increased the less surface area is applied. The second horn is redundant the instant the first makes contact, and may even inhibit penetration distance. Redundancy and symmetry is usually important, hence why a spider has two fangs and likewise Carnivora have two canines, but less is more with puncturing.
Herbivores have teeth adapted for consuming plants. A horse specifically has large incisors and specialised molars, with minimal premolars and no canines (with exception to mature males for fighting) with a massive gap between the incisors and molar row. This poses a problem for eating meat. Canine teeth are needed for grasping, puncturing and tearing meat, hence why mammalian carnivores possess such developed canines. Specialised sheering premolars and molars (look at dogs) are also needed for shearing and slicing meat. A grazing animal can technically chew and digest meat, but not efficiently. Also those big incisors aren’t as useful with meat.
I mentioned before, but I believe a grazing, horse-like ungulate, likely with mating rivalry to evolve horns, is the likely ancestor (disregarding the lack of horns in horses). Let’s say an environmental change inhibits the growth of the ungulate’s main food source, causing them to diversify into omnivorous feeding and developing small animal hunting while still being hunted by their current predator.
Teeth develop, fighting canines present in male horses become expressed in both sexes, ruminant digestive tract de-specialises over time.
As environmental change goes on, the plant life dwindles drastically until the bicorn population reaches near-extinction, causing their predators to starve and die off. Surviving population bounces back slightly due to loss of predation but still dwindles, evolutionary pressure favours their recent predatory adaptions - bicorn becomes obligate carnivore (or more likely opportunistic omnivores).
Along with their new carnivorous niche, their eyes angle forward on their face to stalk prey and allow better depth perception. Their pupils lose the horizontal slit (adapted for scanning the horizon for predators) and adapt a circular or vertical slit better suited for sighting prey.
Because of the less-than-optimal method of hunting that is impaling, the bicorn likely uses sweeping motions to gore prey rather than puncture. This means the horns should be thick and relatively short to deal with the stresses.
Alternatively, if the horns are adapted from male rivalry, perhaps in hunting the pack male/males perform the job of puncturing the prey item while the hornless females herd and pin it down. as well as performing the job of removing the prey item from the male’s horn.
I don’t think this is a particularly likely sequence of events, and it is quite simple, but it’s mostly plausible.