Add read retry policy to EventStoreDb

[DmitryBatalov]

I was testing the ESDB cluster and encountered an issue with failover behavior. When I removed a follower node, the connection did not automatically switch to a healthy follower node and instead continued attempting to connect to the inactive one. I believe this functionality should ideally be handled by the ESDB .NET client, however, as a temporary solution, I would like to implement a retry policy on the Equinox side.