`validate()` should mark `NNNN` alleles for DRB3/4/5

[pbashyal-nmdp]

Typings for DRB3/4/5 can be marked as +ve/-ve with:

Positive: XXXX
Negative: NNNN

eg. DRB3*NNNN.

py-ard currently says they are invalid.

>>> import pyard
>>> ard = pyard.init()
>>> ard.validate('DRB3*NNNN')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
....
pyard.exceptions.InvalidAlleleError: Invalid Allele: DRB3*NNNN is not a valid Allele
>>>

py-ard should treat them as valid alleles.

[mmaiers-nmdp]

That depends. If py-ard is used in a context (WMDA business rules) where they are valid then yes but not if they are not. Similar to tolerating the movement of the expression characters to shortened versions of the alleles, these are WMDA-isms so maybe (like reduce_shortnull) this could be a configurable parameter like allow_wmda_presence_absence

[pbashyal-nmdp]

Thanks for the explanation. Makes sense.