Support serializing state in chunks when saving it on DB

[tbenr]

When we write latest finalized state via setLatestFinalizedState, we serialize the entire state in memory. The state is growing and we should start trying to avoid it.

java.util.concurrent.CompletionException: java.lang.OutOfMemoryError: Java heap space
    at java.base/java.util.concurrent.CompletableFuture.reportJoin(CompletableFuture.java:413)
    at java.base/java.util.concurrent.CompletableFuture.join(CompletableFuture.java:2118)
    at tech.pegasys.teku.storage.server.RetryingStorageUpdateChannel.retry(RetryingStorageUpdateChannel.java:133)
    at tech.pegasys.teku.storage.server.RetryingStorageUpdateChannel.onStorageUpdate(RetryingStorageUpdateChannel.java:86)
    at tech.pegasys.teku.storage.server.CombinedStorageChannelSplitter.onStorageUpdate(CombinedStorageChannelSplitter.java:68)
    at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:103)
    at java.base/java.lang.reflect.Method.invoke(Method.java:580)
    at tech.pegasys.teku.infrastructure.events.DirectEventDeliverer.executeMethod(DirectEventDeliverer.java:74)
    at tech.pegasys.teku.infrastructure.events.DirectEventDeliverer.deliverToWithResponse(DirectEventDeliverer.java:67)
    at tech.pegasys.teku.infrastructure.events.AsyncEventDeliverer.lambda$deliverToWithResponse$1(AsyncEventDeliverer.java:80)
    at tech.pegasys.teku.infrastructure.events.AsyncEventDeliverer$QueueReader.deliverNextEvent(AsyncEventDeliverer.java:125)
    at tech.pegasys.teku.infrastructure.events.AsyncEventDeliverer$QueueReader.run(AsyncEventDeliverer.java:116)
    at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
    at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
    at java.base/java.lang.Thread.run(Thread.java:1583)
Caused by: java.lang.OutOfMemoryError: Java heap space
    at tech.pegasys.teku.infrastructure.ssz.sos.SszByteArrayWriter.<init>(SszByteArrayWriter.java:23)
    at tech.pegasys.teku.infrastructure.ssz.schema.SszType.sszSerializeTree(SszType.java:64)
    at tech.pegasys.teku.infrastructure.ssz.SszData.sszSerialize(SszData.java:58)
    at tech.pegasys.teku.storage.server.kvstore.serialization.BeaconStateSerializer.serialize(BeaconStateSerializer.java:36)
    at tech.pegasys.teku.storage.server.kvstore.serialization.BeaconStateSerializer.serialize(BeaconStateSerializer.java:21)
    at tech.pegasys.teku.storage.server.leveldb.LevelDbTransaction.put(LevelDbTransaction.java:50)
    at tech.pegasys.teku.storage.server.kvstore.dataaccess.CombinedKvStoreDao$V4CombinedUpdater.setLatestFinalizedState(CombinedKvStoreDao.java:576)
    at tech.pegasys.teku.storage.server.kvstore.KvStoreDatabase$$Lambda/0x00007c47949fe8b0.accept(Unknown Source)
    at java.base/java.util.Optional.ifPresent(Optional.java:178)
    at tech.pegasys.teku.storage.server.kvstore.KvStoreDatabase.doUpdate(KvStoreDatabase.java:1127)
    at tech.pegasys.teku.storage.server.kvstore.KvStoreDatabase.update(KvStoreDatabase.java:646)

current mainnet mem reuirement is 246MB

The unterlining DB system (levelDB at least) doesn't support streaming data into DB.
An approach would be to implement a KvStoreVariable that supports having variable stored in chunks, in a similar way to how we deal with "tables" in level db.

an example: https://github.com/andris9/level-stream-access/blob/master/lib/level-stream-access.js

so we could have chunks of (ie 25 MBs) and leverage the int sszSerialize(final SszWriter writer) interface and implementing a chunking KvStoreSerializer and BeaconStateSerializer