It is not impossible to make commercial aircraft "reasonably" hackproof. The problem is that it adds cost and complexity, ESPECIALLY with regards to robust SATCOM systems ($$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$).

But even then there's a big problem. Statistical safety analysis assumes an event will happen with a certain frequency (based on engineering design and analysis). If a HYD pump fails, it's only going to fail on one airplane at any given time and there's no statistical link to other aircraft. So the entire statistical safety system as a whole is only subject to one failure.

But hacking is enemy action... you have to assume that if someone cracked the code, they know they're only going to get one shot to use their hack (before the weakness is pached, or delta force kicks in their door). So they are going for max damage, which means multiple targets all at once, perhaps hundreds or even thousands of airborne aircraft.

Point being, the folks doing safety analysis of a system subject to hacking have to weight the severity of any statistical weakness more heavily to account for the fact that multiple aircraft will be affected all at once.

Another problem... while you might convince yourself that you've designed a statistically hackproof system vis a vis your generic black hat, I doubt you could really have any assurance that any such system would be beyond the reach of a hostile, sophisticated nation-state. While a nation-state would not be inclined to directly perpetrate such a terror attack themselves (any more than they'd intentionally shoot down an airliner), it cannot be ruled out that a hostile nation might provide the tools to terrorists. Unlike most other weapons provided to terrorists or insurgents, hacking tools can be made untraceable.