Terran herbivores' adaptations to terran cellulose and lignin which surrounds each cell in plants was to evolve a wide variety of mechanical and chemical means of breaking down that cellulose and lignin that makes them very different to carnivores, which must merely ingest the food they obtain, which may or may not involve chopping the food into bite-sized chunks.
In the OP's ecology, the line between sessile autotrophic 'plants' and motile heterotrophic 'animals' is far less clearly defined than is the case on Earth. This means that the line between 'herbivores' and 'carnivores' will be similarly blurred. The difference could well be that of speed, the carnivores more adapted to chasing down and catching motile prey while the herbivores will be more adapted to fleeing and/or self-defense.
However, when it comes to eating these soft, fleshy 'plants' with their armour, hard scales, cartilage and thorns, there are a number of obvious adaptations:
The 'herbivores' dentition will likely be very similar to that of a carnivore, in that they would be adapted to slicing up meaty flesh, cutting through tough skin, and scraping flesh from tough, less easily digestable surfaces. This would tend to make a herbivore much more likely to become an opportunistic carnivore. Where plants make themselves abrasive so as to wear down the teeth of the herbivores that eat them, the herbivores may counter with continually growing open-rooted teeth like a horse's molars, or infinitely replaceable teeth like a shark's.
Where the foliage that the herbivores may eat is covered in fine thorns or penetrating hairs, this may be dealt with in the same manner as giraffes which eat thorny acacia foliage deal with the same problem: an armoured mouth and gullet which can tolerate chewing and swallowing sharp, spiky food.
In order to get past thorns and penetrate armoured skins, there are a number of adaptations that could assist:
'Herbivores' heads and any other limbs which may be used to deal with thorny protective measures could develop tough - potentially armoured - skin of their own. If thorns are a problem, tough, horny, smooth surfaces will allow the herbivore to simply push past them. Eyes will be set well back from the mouth to keep them away from the thorns, or they may be armoured themselves.
Thorns may also be dealt with by breaking them off with a protected limb before using other, more sensitive organs and appendages to crack the 'plant' open.
Where a 'plant' has a particularly tough skin, herbivores may develop very powerful jaws in order to simply bite through into the soft interior, or they may develop hammer-like or military-pick-like limbs to respectively smash or puncture the plants' tough rigid or flexible carapaces prior to inserting their relatively slender feeding apparatus into the soft interior.
Limbs may be strong so that when a hole is punctured through the outer armour, the herbivore may use its strength to tear the plant's body open to provide access to the edible interior. An example of this may be the Palorchestids.
Not mentioned by the OP is the possibility that the plants may make their own flesh toxic. This would discourage opportunistic attacks by carnivores and unspecialised herbivores which may be poisoned, but some herbivores would evolve to be able to neutralize any toxins that the plants they feed upon may create. However, this may increase the metabolic cost of feeding upon these toxic plants, a problem that tends to lead to a large size and a leisurely lifestyle in order to reduce energy consumption per unit mass, large animals having a lower basal metabolic rate to smaller animals.
Many of these adaptations to herbivory in this environment would also lend themselves to carnivory, so true exclusive 'herbivores' and 'carnivores' may not actually exist, but instead each species may lie somewhere along the scale of omnivory between the two extremes.
Also not mentioned by the OP is the possibility of symbiotic relationships between 'plants' and certain 'animals' where the animals protect the plant in exchange for some incentive such as food and/or shelter. Eating such a plant would require dealing with both the plant's innate defences and its symbiotic defenders, though an animal evolved to do that may well simply eat both plant and its o(symbiotic defenders.