Add policies with higher replication factor

FDB provides multiple replication/locality configurations (`single`, `double`, `triple`, `three-data-hall` etc). Most of these (I am ignoring the datacenter modes as I consider those being HA solutions) allow for up to two machine failures without any data loss.

Amazon Aurora, which is what people are getting used to, seems to be slightly better than this:

> Amazon Aurora is designed to transparently handle the loss of up to two copies of data without affecting database write availability and up to three copies without affecting read availability.

I would like to be able to provide a SLA that is at least as strong as whatever Aurora provides. Allowing for higher replication factor would also give us better read throughput.

Start by reviewing the existing `single`, `double`, `triple`, and `three-data-hall` replication/locality configurations and the surrounding replication design. The issue names no files or tests; done would require an agreed design and implementation for higher replication factors that meets the stated write/read availability and read-throughput goals.

Written by the indexing model from the issue text.