Not supposed to is not the same thing as was designed to. If the car stock can handle its rated performance levels with normal failure rates, its not really the fault of the manufacturer that it starts to fail as soon as people upgrade it. Some engines through the years have been designed to be very upgrade friendly through specific design choices or just being over-built in general. You find that a lot in diesel engines and many older blocks that were just built like brick shit houses cause it was easier to overbuild than not.
The crankwalk issue that the EA888 apparently has in some areas is the 180-degree thrust bearing (which is a very common design these days) isn't as strong for upgraded clutches as past 360-degree bearings. Some failures are sprinkled in on stock cars, but if you look at the entire production versus those stock cars, its barely a blip. Manufacturing flaws could easily encompass those, and be totally normal versus like model vehicles.
I think the tuning market in general is focusing their energy in the wrong area. Crankwalk is a problem on upgraded clutches, the issue is a small bearing, not a faulty bearing. Performance junkies will spend countless dollars on upgraded parts, sweet looking exhaust systems, ECU tunes, etc etc... why not budget to include machine shop time to machine a full diameter bearing? The problem has been solved for the most part by many independent outfits as we've seen linked to across the various threads. Machine a bearing cap, include a larger or additional thrust washer, reinstall. A sub $500 solution is already out there, which a competent machine shop could handle. If the aftermarket sees this as a large enough opportunity, maybe someone should start a commercial process of upgrading existing crank bearing caps (ship yours in, they machine it and ship it back).
These are the same guys that swap in metal oil pans vs plastic, upgrade turbos, hell... upgrade the clutch itself in the first place. Why can't they also account for an upgraded bearing in the process?