bug: L1 subgraph incorrectly calculating retryableTicketId for bridge txs

[tmigone]

There isn't a clear pattern yet but sometimes the L1 subgraph calculates the ticket id incorrectly, which leads to the monitoring system firing false positives.

Example

• L1 tx: https://etherscan.io/tx/0x4de233b4503540d9a0111bd9e51a01df93fbeaebe46d2efcc2d5bc7c1bff3f9d
• L2 ticket creation tx: https://arbiscan.io/tx/0xb0192474e324a190bfbcf5e3fecaedff1a6e52bc9eb8e32a960d68a4eb0a6d17
• L2 ticket auto-redeem tx: https://arbiscan.io/tx/0x90ca20e639d6e30001a1b4f5d776aac1959d1d0cb29e5fea88173d01af8cfe24
• L1 subgraph calculated ticket id: 0x88c11cd46a8f310cce8d7bdf5795968e6ca117f23496e8484f53ff689bcf99a8
• Actual ticket id: 0xb0192474e324a190bfbcf5e3fecaedff1a6e52bc9eb8e32a960d68a4eb0a6d17