It's actually the other way around. We first need a solution for the quadratic hashing time problem before we can think about a blocksize increase. Otherwise, the danger of spam is too big.

I know, FlexTrans does also provide a solution. But as with Segwit, we would need first a pretty long timeframe to allow nodes to adopt the new transaction format. And when the block size increase goes live, we would need a restriction to legacy transactions because otherwise spammers could simply use them to continue spamming the network with transactions that would crash smaller full nodes.