Follow Strelka BP and use Manta small indels candidates as input

[maxulysse]

new PR to replace #467
This is deactivated by default.
So it should be safe to merge if we want to try this new feature

[maxulysse]

I think I can add this option for Germline as well, so do not merge yet.

[maxulysse]

So bad idea to try to implement for germline, I should have read further.

The germline caller already includes its own assembly step, so indel candidate input from a tool like Manta is not recommended in this case (in testing, Manta candidate indels have been found to provide no benefit and in some cases may complicate the accurate resolution of a locus).

At least I read it before starting to do some work on it.

[szilvajuhos]

Looks OK

[szilvajuhos]

Eventually we could check on real data whether it improves indel calls.

[maxulysse]

Of course, that why for the moment it's not working by default, but I hope to definitively use this functionality.
We can provide both VCFs from Strelka, for the moment, we now do have two different directories, so it'll be easy to check it out