How did it "have to be?" No, it really didn't. If it doesn't meet the definition of being a classified tropical system, it simply isn't.

What the Canadian model is showing is baroclinic development in 2 and a half days and beyond -- in other words, not of 93L -- and not of a tropical nature.

http://moe.met.fsu.edu/cyclonephase/cmcglb/fcst/archive/06110412/100.html
and
http://moe.met.fsu.edu/cyclonephase/cmcglb/fcst/archive/06110412/124.html

That is more typical of a warm-seclusion development with a broadening wind field.