Prefix compress read range results

Generally all of the key-value pairs returned by a read range will share a common prefix. When serializing read range results which should take out this common prefix to reduce the amount of bytes sent across the network.

The issue provides no file, test, or entry point to begin from. First identify where read range results are serialized and how their shared key prefix is represented across the network. Done means preserving result semantics while reducing the serialized byte count, with coverage for prefix-sharing ranges.

Written by the indexing model from the issue text.