The most widespread method of stopping spam in any form is to use flags to train an machine learning algorithm that picks out particular patterns in spam. These algorithms don't have political agendas, they just pick out common characteristics of spam. It is therefore entirely unsurprising that any such spam detector would learn to only flag women, and therefore to be expected that a bug in that algorithm which generated false positives would affect only women's profiles.
The reason you are downvoted is not because you are trying to highlight women's issues (I got 50+ karma for calling out another comment as sexist against women a few weeks ago), but because you were quick to imply a sexist motive for behavior that most any hacker would have immediately recognized as a natural consequence of how poorly-performing spam-detectors behave.