So first, ignore the narrative from the CogAT report. That's boilerplate, generated automatically by the interpretive software, in response to the difference between verbal and nonverbal/quantitative. Even on this test, his verbal is well strong enough not to need supplementation.
Secondly, it's not impossible that this is a realistic verbal score. It's still in the top of the high average range, and compatible with being in the top 1-2 students in his class in reading/writing.
Thirdly, although he reads widely for pleasure in English, he probably spends much less time than a similar-ability child of his age does on informational text/academic reading in English, which may have mildly depressed his verbal score. (You report that he is reading almost solely literary text of a very specific genre.)
And on bubbling: so that's not necessarily a skill closely correlated to overall intelligence! Actually, it also makes a little bit of sense when you consider that historically, processing speed (which is typically assessed in ways that overlap significantly with fine motor speed) has been his area of personal weakness. It appears he overlooked some bubbles on his way through Q and N. I suppose it's also not impossible that his lower-than-you-expected V score involves getting off-track at some point in bubbling, but you'd need to have someone handscore his response booklet (I'm assuming this was paper, since you mention bubbling at all) with that specific type of error in mind, so they can see if there's a series of answers anywhere along the way that appears to be shifted.
In conclusion: not worried.
