Pick multiport lanes by inside flow hash, not routine index

Lanes were collapsing onto lane 0 and staying there. Nebula writes a
peer's inbound packets to the tun queue matching the socket they arrived
on, and tun_flow_update teaches the kernel to steer that flow's outbound
packets to the same queue, preferring what it learned over the hash. So
while a tunnel's lanes were still down -- which is every tunnel, for its
first moments -- all traffic arrived on socket 0, pinning every flow to
queue 0 on both hosts for as long as it stayed busy. With the lane
following the routine, that meant lane 0 forever.

Two changes break the loop:

Choose the lane from the packet's own 5-tuple hash, so lane spread no
longer depends on tun steering at all. The hash is symmetric, and the
high-addressed side rotates its choice by the low side's port offset, so
a flow's two directions land on partner lanes with exact reverse
4-tuples and each arrives through the conntrack entry the other's probe
opened.

Seed every lane as demanded at lane-set creation, so the first traffic
tick probes them all at once rather than waiting for a flow to hash onto
each. Lanes have to be up before the flows are, not after.

A routine may now send on any lane, so it holds a SendBatch per lane,
built lazily and all borrowing one shared arena -- the slab is the
expensive part, and a batch per (routine, lane) would otherwise cost
hundreds of megabytes. full() counts across the batches so the total
outstanding stays bounded as before.

That also means several routines can write to one socket, which was
already true for base traffic under multiport: the linux batchWriter's
sendmmsg scratch had no lock. Take a mutex there, once per flush.

udp/udp_linux_writebatch.go

@@ -12,6 +12,7 @@ import (
 	"net/netip"
 	"strconv"
 	"strings"
+	"sync"
 	"unsafe"
 
 	"golang.org/x/sys/unix"
@@ -19,8 +20,14 @@ import (
 
 // batchWriter owns the sendmmsg(2)/UDP-GSO transmit path for a StdConn: the
 // scratch WriteBatch packs mmsghdr entries into, plus the GSO capability
-// state probed at socket creation. Each queue has its own StdConn and
-// batchWriter, so no locking is needed.
+// state probed at socket creation.
+//
+// One socket can have several senders: under multiport every routine writes its
+// base-session traffic to socket 0, and any routine may write to any lane
+// socket, since a packet's lane comes from its flow rather than from the routine
+// that read it. The scratch is per socket, not per sender, so mu serializes
+// packing and draining. It is taken once per flush — a syscall's worth of work —
+// and only contends when two routines flush the same socket at the same instant.
 //
 // Terminology, smallest to largest:
 //
@@ -42,6 +49,10 @@ type batchWriter struct {
 	fd   int
 	isV4 bool
 
+	// mu guards everything below it: the scratch, and the GSO state WriteBatch
+	// can clear.
+	mu sync.Mutex
+
 	// UDP GSO (sendmsg with UDP_SEGMENT cmsg) support, probed once at
 	// socket creation and cleared by WriteBatch if the kernel later rejects
 	// a GSO send (the setsockopt probe cannot see per-route limitations).
@@ -196,6 +207,9 @@ func (w *batchWriter) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, er
 		return 0, fmt.Errorf("WriteBatch: len(bufs)=%d != len(addrs)=%d", len(bufs), len(addrs))
 	}
 
+	w.mu.Lock()
+	defer w.mu.Unlock()
+
 	// A destination the kernel rejects results in us dropping that entry (one packet, or one same-destination GSO run).
 	// We count what actually made it out rather than returning an error.
 	written := 0