Incorrectly identifies a long-standing occasional crash as being a new regression

[jfkthame]

See https://bugzilla.mozilla.org/show_bug.cgi?id=1517767.

This bug was filed based on a single recent crash report, and marked as a regression; but in fact it has happened sporadically for months: searching for the signature on crash-stats shows 56 reports in the past 6 months, spread fairly evenly with 0-4 per day.

So this clearly isn't a new regression, the tracking flags (earlier versions being marked "unaffected") are wrong, and the suggested regressing bug is irrelevant.

[sylvestre]

I am pretty sure this was calixte and not the bot filling this bug ;)

[jfkthame]

Oh, OK... I just saw that it was marked as blocking the clouseau meta-bug and so I assumed it was the bot's work, but maybe not.