This is the minimal necessary change to make the next commit work.

This change significantly improves the performance of RLPx message reads
and writes. In the previous implementation, reading and writing of
message frames performed multiple reads and writes on the underlying
network connection, and allocated a new []byte buffer for every read.

In the new implementation, reads and writes re-use buffers, and perform
much fewer system calls on the underlying connection. This effectively
doubles the theoretically achievable throughput on a single connection,
as shown by the benchmark result:

    name             old speed      new speed       delta
    Throughput-8     66.5MB/s ± 0%  139.7MB/s ± 1%  +110.16%  (p=0.001 n=7+7)

The change also removes support for the legacy, pre-EIP-8 handshake encoding.
As of May 2021, no actively maintained client sends this format.

The snappy functions only check the length of the destination buffer and
ignore its capacity. When message sizes alternated between large and
small, we would still allocate a new buffer for every other message.
This fixes it.