I get what you're saying, but I don't think that really fits the criteria. Yes, miners implicitly agreed that the coin would fork, but there was just a block height set by the dev that it would happen on, and I think generally that code was included in the original release. There hasn't been, as far as I'm aware, a coin that began its life as PoW, then at some point a new version was released and it switched to PoS after X out the last Y blocks were mined on the new version.
Edit: I think the distinction here is that of a consensus/community approved hardfork and one that is consensus driven. That is, in the approved case the hardfork client is released and block height announced, users and exchanges switch over at the behest of developers, so if miners don't then they're left mining on a worthless chain. Of course, it's possible users and/or exchanges refuse to switch, in which case the fork would be denied, but this doesn't happen much afaik in crypto world. In the consensus driven case the changes are not implemented until a verifiable percentage of blocks have been mined and recorded on the blockchain with the new version number.