A cache for a quicker initial catch-up with bitcoind

On [Lamp](https://github.com/lvaccaro/lamp) the startup sync with `bitcoind` takes approximatively 2 orders of magnitudes more time using my Esplora instance or bs.info than using a local bitcoind RPC.

I think we should detect by how much block is `lightningd` far from the tip and cache blocks from Esplora so that the biggest above bottleneck (bandwidth) is compensated.

This is halfly a self-assignment and remainder but if someone wants to work on this feel free (just comment here).

Begin by tracing the startup synchronization between lightningd and the Esplora or bs.info backend, using local bitcoind RPC as the performance comparison. Done means an agreed cache design that reduces the initial catch-up bandwidth bottleneck while keeping the chain-tip gap and fetched blocks correct.

Written by the indexing model from the issue text.