Fixes #400

Checklist

• Added validaton to any changes in the parser API.
• Added tests covering the parsing behaviour changes.
• Linked to any relevant issues this will close.
• Tested the output using the demo.

Empty alt text (<img alt="" .../>) is used for implied name properties but the spec specifically mentions not using an empty value.

Also changed the behavior to use impliedTextContent instead of textContent when parsing p-* otherwise it adds the images src when parsing.

Example input covered by new behaviour

<a class="p-author h-card u-url" href="/">
  Author
  <img class="u-photo" src="/photo.jpg" alt="" />
</a>

Example output from new behaviour

{
  "items": [
    {
      "type": ["h-card"],
      "properties": {
        "photo": ["http://example.com/photo.jpg"],
        "name": ["Author"]
      }
    }
  ]
}