That's my reading of the prosecutor's decision as well: her past - including an improbable number of sexual assault allegations several of which were never proven - brings her credibility into question, making it the legitimate target of cross-examination by the defence. That cross-examination would predictably be very difficult for the victim, which is a major factor in the prosecution deciding to opt for a plea deal. The corroborative evidence suggests it's more likely her story was, in fact, true.
I disagree with respect to this not being a race/religion issue. Legally speaking, the issue of the man's race/religion is central to the decision: it was his deception on this point that allowed for a conviction of rape by deception. The information contained in the testimony of the victim is, as far as the law is concerned, irrelevant.
I also wouldn't read too much into the media's initial reporting of this case. Media outlets pay people to scan over legal decisions looking for anything topical or newsworthy - they don't usually have people in the courtroom for the entire trial process (and almost certainly didn't in this case). Anyone reading the court's decision wouldn't know the background of the case - they'd have to go to the transcript for that. So I'm not at all surprised the initial story got printed without this new information.
What concerns me more about the media's reaction is why the new information hasn't been more widely published in media outlets that carried the initial story.
Edit: Here's a nice piece by Canadian-Israeli freelancer
Lisa Goldman that summarizes things nicely.
And an interesting excerpt from her article raises more concerns about the scope of the "rape by deception" law:
There were two precedents of Israeli Jewish men convicted of rape by deception; but in both those cases, the men were convicted of lying about their socio-economic status. Never had a man been convicted of rape by deception for claiming he was a Jew when he was in fact an Arab.
The article links to articles describing the two precedent cases, which are interesting in their own right.