following #101 and #102, I took the commits from #103 and finished them:
Now, in any given row, each sample can have different ploidy.
As the ploidy affects the cardinality of fields like PL, with this PR each sample can have different cardinality in those fields.
I kept PL Number=G even in 4.1 and 4.2, because the problem (different samples needed different field cardinality) is fixed by the line above this one.
following #101 and #102, I took the commits from #103 and finished them: