Improve compression robustness - Part 1

[tudor-malene]

Why this change is needed

The sequencer creates and compresses rollups relative to an l1 block it thinks is canonical at that time.
If it then turns out that block is a reorg and some validators don't have it, they can't decompress.

The solution is:

1. for the sequencer to recompress and resubmit the rollup relative to the new l1 head, when there is a reorg on a block that was used for a rollup.
2. To minimise the chance there will be a delay added. You never include in the rollup the batches from the current l1 head.

This PR implements 2. and fixes the headers

What changes were made as part of this PR

• remove unused field from the rollup header and replace with the : CompressionL1Head
• modify the rollup creation to include the delay
• remove extra rollup compression

PR checks pre-merging

Please indicate below by ticking the checkbox that you have read and performed the required
PR checks

• PR checks reviewed and performed