maybe @garlonicon can tell you if he went through with it
He definitely didn't. His mining software left some "extranonce-like" traces in the blockchain, so you can easily check, that he didn't mine any testnet blocks in 2025. Of course you can try to PM him, but I don't think he has in store some testnet3/testnet4 secrets, which he didn't share publicly.
And also, when I talked with him recently, he gave me some hints about signet, so I think he is now more focused on hacking signets than testnets (and I guess it is a good choice, if CPU mining will be blocked in testnet5, and if testnet3 and testnet4 will be abandoned by Core developers).
That took 1 second.
Wrong. Sending that data to your node took a short amount of time. But mining was slower, and your node just ignored blocks on the same (or lower) height.
If you want to see it, you can run two nodes on localhost, and check, what happens between miners with similar hashrate (which is the case here, because every CPU-mined block brings the same chainwork, no matter who produced it).
It seems to me that the only chance to stop what is happening now on testnet is to lead and distribute for free, thereby devaluing the test coin sales market.
You think, that traders really care about technically running any nodes? As long as centralized trading works, they will trade. Testnet4 can even be halted every 2016 blocks, and nobody cares (because then, all miners are forced to work with real ASIC difficulty for a single block).
There existed some CPU-mined altcoins, where validating the chain was harder, than mining the next block. And it didn't affect trading at all, and people shared already validated database in a "don't verify, trust" way.