Citus optimize multi-shard range search queries

[xin-hedera]

Problem

Rest API performance tests show citus performs quite worse for multi-shard range search queries and it doesn't scale well with increasing number of virtual users

Solution

There are multiple approaches we can experiment and they are not mutually exclusive

- [x] #5438
- [x] reduce the number of shards from the default 32 to 16
- [x] add readonly coordinator nodes and spread the read queries
- [x] Add pgbouncer to each node
- [x] add worker nodes / increase the max connections allowed between coordinators and worker nodes
- [ ] #5621
- [ ] #5622 
- [ ] #8287
- [ ] https://github.com/hashgraph/hedera-mirror-node/issues/5230
- [ ] https://github.com/hashgraph/hedera-mirror-node/issues/5229
- [ ] https://github.com/hashgraph/hedera-mirror-node/issues/5234

Run the REST performance tests to check if performance improves

Alternatives

No response