Why do the first three transactions need to be sent to the execution layer's memory pool? It seems that on the fourth time, all four transactions will be included in the block through mev-boost's force inclusion.
We send transactions to the memory pool so that they can be included earlier. At the fourth time (the last L2 slot, 3s each), we force all transactions that have not been included up to that point to be included.
https://github.com/NethermindEth/Taiko-Preconf-AVS/blob/ec553f79333a3a22494a596c4f0d2e11cad6da82/Node/src/ethereum_l1/execution_layer.rs#L265
Why do the first three transactions need to be sent to the execution layer's memory pool? It seems that on the fourth time, all four transactions will be included in the block through mev-boost's force inclusion.