reduce reprocess queue/channel sizes (#3239)

## Issue Addressed

Reduces the effect of late blocks on overall node buildup

## Proposed Changes

change the capacity of the channels used to send work for reprocessing in the beacon processor, and to send back to the main processor task, to be 75% of the capacity of the channel for receiving new events

## Additional Info

The issues we've seen suggest we should still evaluate node performance under stress, with late blocks being a big factor. 
Other changes that could help: 
1. right now we have a cap for queued attestations for reprocessing that applies to the sum of aggregated and unaggregated attestations. We could consider adding a separate cap that favors aggregated ones.
2. solving #2848

beacon_node/network/src/beacon_processor/mod.rs

@@ -89,7 +89,7 @@ pub const MAX_WORK_EVENT_QUEUE_LEN: usize = 16_384;
 const MAX_IDLE_QUEUE_LEN: usize = 16_384;
 
 /// The maximum size of the channel for re-processing work events.
-const MAX_SCHEDULED_WORK_QUEUE_LEN: usize = 16_384;
+const MAX_SCHEDULED_WORK_QUEUE_LEN: usize = 3 * MAX_WORK_EVENT_QUEUE_LEN / 4;
 
 /// The maximum number of queued `Attestation` objects that will be stored before we start dropping
 /// them.