Description of problem: 
diagrams with unclear differentiation between captions and the paragraph that follows.
  
Actual results:
the current XML looks something like this....

<mediaobject>
	<imageobject>
		<imagedata fileref="images/producer-consumer.png" format="PNG" scalefit="1" width="444" />
	</imageobject>
	<textobject>
		<phrase>The AMQP 0-10 Model</phrase>
	</textobject>
	<caption> <para>
		The AMQP 0-10 Model: Message producers write messages to 	exchanges, message consumers read messages from queues
	</para>
	</caption>
</mediaobject>

Expected results:
XML should look more like this.

<figure id="AMQP_0-10_Model">
	<title>The AMQP 0-10 Model</title>
	<mediaobject>
		<imageobject>
			<imagedata fileref="images/producer-consumer.png" format="PNG" scalefit="1" width="444" />
		</imageobject>
	<textobject>
		<phrase>The AMQP 0-10 Model: Message producers write messages to exchanges, message consumers read messages from queues</phrase>
	</textobject>
	</mediaobject>
</figure>

Additional info:

Existing <textobject> to become new <title> and existing <caption> to become <textobject>. Textobject contents aren't shown in docs when the image is so check that the content in them won't be missed/is covered in paragraphs around image.