As VIP developer I want (VCF) tests to compare test results with a truth set so we have better testing that indicate how many variants were properly classified

[MatthieuBeukers]

Objective
We can already improve several (mostly VCF) tests for which we have a truth set (vkgl, mtdna, jvar, fazzini). Instead of looking at the number of lines we could intersect the VIP results with the truth set to assess how well the test has performed.
Tests that are candidates for example:

• fastq/mtdna_fazzini_gs
• vcf/jvar_blb
• vcf/jvar_plp
• vcf/mtdna_blb
• vcf/mtdna_plp
• vcf/vkgl_lb
• vcf/vkgl_lp

Implementation suggestion (optional)
At the end of the test we could use bcftools isec to intersect the resulting vcf with the test truth set and count the number of shared variants.