"Wrote code" could include a wget cron job. "Bypass company's security" includes literally any activity done without company approval or authorization, including printing out files and putting them in his briefcase and walking out the building without announcing the fact.
No one is really disputing that his alleged acts were "treacherous" to the company, or that they "caused damage to the company's interests" -- that would cover every conceivable form of whistleblowing.
At this point, I don't see what it matters whether it was a "real" hack or not (though it's ironic you mention social engineering, since for too long that has been ignored as a real attack vector). He took info and disclosed it without company approval, now it's up to the courts to decide if that was legitimate and protected whistleblowing.
As to your last sentence, I think it would be difficult to argue that automated and ongoing bulk data exports could ever fit with any definition of whistle-blowing. If he had legitimate concerns, and evidence of those concerns existed in electronic form, he could have simply walked the specific evidence out of the building on a thumb drive.
That being said, we don't know exactly what data was exported, so any speculation (including my own) is rather pointless...
The real issue is whether he is a legitimate whistleblower not the acts that he did in order to provide the data to journalists.
He wrote code that was explicitly designed to bypass the company's security and treacherously act in a way that caused damage to the company's interests, then he inserted it using other people's security credentials. The only elements missing between what he did and what a "real" hacker might have done is breach a technical security barrier rather than be a trusted employee.