Well, you give a mission plan of one year, and design a rover to fit that plan. And because you don't want to fail, you engineer everything to have a tiny failure rate in the first year. That also makes them likely to work for many more years, but that's just a side effect. You have only explicitly designed and spent money on that first year. Taking funding from that first year to add a system that only becomes useful in year three is a completely different matter, and likely much harder to greenlight.

Also, something not explicitly stated in this thread is that most of the most important science happens in the first year the rover / lander is there on Mars. You really want that first year to be successful, and anything beyond that is bonus.

Sure, but you also learn from experience. If it turns out dust on panels is usually the limiting factor, and the rover otherwise stay in good shape well beyond expectations, a mitigation for the dust may be an effective solution in reliably operation lifetime and thus mission scope.

Every component runs into issues, look at how many rover’s wheels had issues before the panels failed.