Improve eth1 block sync (#2008)

## Issue Addressed

NA

## Proposed Changes

- Log about eth1 whilst waiting for genesis.
- For the block and deposit caches, update them after each download instead of when *all* downloads are complete.
  - This prevents the case where a single timeout error can cause us to drop *all* previously download blocks/deposits.
- Set `max_log_requests_per_update` to avoid timeouts due to very large log counts in a response.
- Set `max_blocks_per_update` to prevent a single update of the block cache to download an unreasonable number of blocks.
  - This shouldn't have any affect in normal use, it's just a safe-guard against bugs.
- Increase the timeout for eth1 calls from 15s to 60s, as per @pawanjay176's experience with Infura.

## Additional Info

NA

beacon_node/http_api/src/lib.rs

@@ -2075,9 +2075,7 @@ pub fn serve<T: BeaconChainTypes>(
                 let head_info = chain
                     .head_info()
                     .map_err(warp_utils::reject::beacon_chain_error)?;
-                let current_slot = chain
-                    .slot()
-                    .map_err(warp_utils::reject::beacon_chain_error)?;
+                let current_slot_opt = chain.slot().ok();
 
                 chain
                     .eth1_chain
@@ -2088,7 +2086,7 @@ pub fn serve<T: BeaconChainTypes>(
                         )
                     })
                     .and_then(|eth1| {
-                        eth1.sync_status(head_info.genesis_time, current_slot, &chain.spec)
+                        eth1.sync_status(head_info.genesis_time, current_slot_opt, &chain.spec)
                             .ok_or_else(|| {
                                 warp_utils::reject::custom_server_error(
                                     "Unable to determine Eth1 sync status".to_string(),