Pissing Attack / adversarial blocks attack

How do the ones of you familiar with the protocol and game theory suggest dealing with potential ’empty block attacks’ (a majority of miners collaborate to slow down Bitcoin transactions by producing empty blocks), or even more so, with ‘adversarial block attacks’ (a majority of miners collaborate to sabotage the network by sending blocks containing only a number of legit transactions), also known as ‘pissing attacks’?

Empty blocks are easier to identify and don’t contain valid UTXO’s, so I suppose the protocol just ‘skipping’ them wouldn’t do any damage (although, if it wouldn’t, why wasn’t that implemented already?).

But adversarial blocks? Any acknowledgment that the nodes can simply identify and zap transactions that they don’t like would also be an acknowledgment that Bitcoin is not censorship proof, and open up a wormhole of potential consequences, while very possibly still not allowing to stop such attacks at an early enough stage to prevent lasting damage in trust and credibility (not to speak of price/stored value).

The risk for nation-state attacks is still relatively low, but might have to be dealt with at some point.