Write race condition while migrating/importing a slot

[hengyoush]

Bug Report

When migrating a slot, setting the slot to a "migrating" state may occasionally trigger MOVED errors when accessing keys in that slot. I've noticed that Issue #1218 on the Lettuce GitHub repository (#1218) describes a similar problem. However, the fix for this issue appears ineffective because in DefaultEndpoint.writeToChannelAndFlush, even when a command list is passed in, channelWrite still sends each command individually. This implementation may lead to a write race condition.

private void writeToChannelAndFlush(Collection<? extends RedisCommand<?, ?, ?>> commands) {
    QUEUE_SIZE.addAndGet(this, commands.size());
    
    if (reliability == Reliability.AT_MOST_ONCE) {
        // Cancel on exceptions and remove from queue (no housekeeping)
        for (RedisCommand<?, ?, ?> command : commands) {
            channelWrite(command).addListener(AtMostOnceWriteListener.newInstance(this, command));
        }
    }
    
    if (reliability == Reliability.AT_LEAST_ONCE) {
        // Commands remain in queue for reconnect retries
        for (RedisCommand<?, ?, ?> command : commands) {
            channelWrite(command).addListener(RetryListener.newInstance(this, command));
        }
    }
    
    channelFlush();
}

Expected behavior/code

Environment

• Lettuce version(s): [6.3.2.RELEASE]
• Redis version: [6.2.13]

If this is indeed a bug, i would like to send a pr.