[AMQP Python] remove redundant sequence body normalization

[swathipil]

Currently, when converting an AmqpAnnotatedMessage with a sequence body, we "normalize" by wrapping the sequence body in a list. Then we iterate through the list when encoding the sequence body.

As per this discussion, we don't need to "normalize" since that's redundant.
TODO:

1. Remove the "normalize_sequence_body" in the AmqpAnnotatedMessage constructor so that it's just:

elif "sequence_body" in kwargs:
            self._sequence_body = kwargs.get("sequence_body")
            self._body_type = AmqpMessageBodyType.SEQUENCE

2. Remove the loop from the _pyamqp/_encode.py file when encoding sequence body so that it looks like:

if payload[6]:  # sequence
        encode_value(
            output,
            {
                TYPE: AMQPTypes.described,
                VALUE: (
                    {TYPE: AMQPTypes.ulong, VALUE: 0x00000076},
                    {TYPE: None, VALUE: payload[6]},
                ),
            },
        )