One of the great benefits that being in touch with the inner moralizer can have is that can warn you about how what you say will be interpreted by another. It would probably recommend against speaking your first paragraph, for example.

If it works that way, I don't want it. My relationship with X has no value to me if the relevant truths cannot be told, and so far as I can tell that first paragraph was both true and relevant at the time.

Now if that had been a coworker with whom I needed ongoing practical cooperation, I would have made some minimal polite response just like I make minimal polite responses to statements about who is winning American Idol.

...But it is also a different underlying mistake than the typical ad hominem.

Okay, there might be some detailed definition of ad hominem that doesn't exactly match the mistake you described. I presently fail to see how the difference is important. The purpose of both ad hominem and your offered interpretation is to use emotional manipulation to get the target (me in this example) to shut up. Would I benefit in some way from making a distinction between the fallacy you are describing and ad hominem?