Is there any good reason why so many miners are apparently mining it?
They must have an incentive. I can't really believe that paying the hosting of a faucet website plus the mining costs are less than the ad revenue, but they must have a profit somehow, otherwise they wouldn't mine.
That profit may not be translated into money.Interesting thought; however I'm not sure what else could motivate them.. Do you have anything particular in mind (even if just pure speculation)?
The moment when tBTC is sold and bitcoin community know it, there will be discussion to abandon testnet 3 network and create testnet 4 network

There is "testnet 4" called signet that is used now, because of blockstorms and other testnet issues. Getting large amount of signet coins is easier than testnet coins, so testing in signet should be easier. Also, the only reason why someone would need huge amounts of that coins is just for paying minimum transaction fees. But assuming zero satoshi outputs and one satoshi per byte fees, that means every 0.01 tBTC is enough to create 1 MB of test transactions. That means 2.5 tBTC is enough to send 250 MB of test transactions. But then, does someone really need to put 250 MB on-chain? Or maybe that amount of data could be lowered by using Lightning Network inside testnet?
True, I would for sure use
signet e.g. for developing applications that run on Bitcoin. When I last did something like that, signet wasn't out yet, so I haven't tried it myself so far - not sure if it has some limitation that would incentivize using testnet instead.
Regarding the name: I find it good that it's not called testnet4, because it's actually not

It has a few notable changes and is not just a newly setup 'testnet' with new genesis block, like testnet 1 through 3 were.