p2p/rlpx: reduce allocation and syscalls (#22899)

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 doubles the
theoretically achievable throughput on a single connection, as shown by
the benchmark result:

    name             old speed      new speed       delta
    Throughput-8     70.3MB/s ± 0%  155.4MB/s ± 0%  +121.11%  (p=0.000 n=9+8)

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

p2p/transport.go

@@ -25,6 +25,7 @@ import (
 	"sync"
 	"time"
 
+	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/common/bitutil"
 	"github.com/ethereum/go-ethereum/metrics"
 	"github.com/ethereum/go-ethereum/p2p/rlpx"
@@ -62,6 +63,10 @@ func (t *rlpxTransport) ReadMsg() (Msg, error) {
 	t.conn.SetReadDeadline(time.Now().Add(frameReadTimeout))
 	code, data, wireSize, err := t.conn.Read()
 	if err == nil {
+		// Protocol messages are dispatched to subprotocol handlers asynchronously,
+		// but package rlpx may reuse the returned 'data' buffer on the next call
+		// to Read. Copy the message data to avoid this being an issue.
+		data = common.CopyBytes(data)
 		msg = Msg{
 			ReceivedAt: time.Now(),
 			Code:       code,