1992-custom-snapshot-intervals

[Cmdv]

Fix: Reduce snapshot frequency during initial sync

Fixes issue #1992 by making ledger state snapshot intervals configurable and significantly reducing snapshot frequency during initial sync to improve performance.

Changes:

• Added snapshot_interval configuration with following (default: 500) and lagging (default: 100,000, previously 10,000) parameters
• Fixed bug in timeToSnapshot function that was ignoring the lagging interval (previously hardcoded to False)
• When near tip, snapshots continue every 500 blocks for quick rollback recovery

New configuration:

 {
   "snapshot_interval": {
     "following": 500,
     "lagging": 100000
   }
 }

[kderme]

Note that this actually increases the frequency of snapshots instead of reducing it. As the ticket suggests

During syncing dbsync takes a snapshot every epoch. ... We should 
take them much more rarely, at least for the majority of the syncing. 
When dbsync reaches the tip of the chain, for example it's within 3-5
epochs from the tip, it can continue with snapshots for every epoch.

These lines could be relevant https://github.com/IntersectMBO/cardano-db-sync/pull/2006/files#diff-414b9fb12c496ccef0f8796908e01b3c282a0d1db1434da6d53756104e524f21R333