fix duplicate heatmap for kraken

[jchorl]

The kraken module takes the top 5 taxons and shows duplication across samples.

The problem is when some samples don't include those taxons. The heatmap "collapses left" because of how the dict is turned into a list:

This means the duplication amount shown on the second row is actually for the wrong taxon!

Really, we should fill in None (or 0?) when a taxon isn't present in a sample. After fixing:

• This comment contains a description of changes (with reason)
• CHANGELOG.md has been updated

[ewels]

Nice, improves the MultiQC_TestData too:

to this:

[ewels]

In fact, I don't think that this heatmap needs to be square.. Tacked a minor config improvement on to your PR 👍🏻