[Mempool spam] Should we as developers reject non-standard Taproot transactions from full nodes?

On the bitcoin-dev mailing list, a discussion took place regarding the recent congestion of the Bitcoin mempool due to side projects like BRC-20. Ali Sherief highlighted the threat that such tokens pose to the smooth use of the Bitcoin network, as real transactions are being priced out, and suggested taking action in the form of BIPs and/or commits into the Bitcoin Core codebase to curtail the unintended consequences of BIP 342. Luke Dashjr agreed that action should have been taken months ago and proposed extending existing spam filters to Taproot transactions or implementing a more narrow approach like OP_RETURN. Jaroslaw chimed in to highlight the real reason for feeling "something must be done", which he believed was the $30 fee per transaction, necessary to compensate for the lack of block subsidy long term. He paraphrased some quotes from the discussion and emphasized the need to find a solution for the long-term fees problem that fits everyone's common ground. He also mentioned delaying halving in case of a 4-year-long network difficulty regression situation. While some may criticize this proposed solution, it is important to ensure that this kind of congestion can never happen again using Taproot, as the community indirectly allowed it to happen, according to Ali Sherief.