My advice here is worth nothing, but I'd probably be working in paranoid mode at this point. I'd consider it a failed proof test, and then work to figure exactly what failed and how. If the manufacturer breeched the barrel, I'd be talking with them and not necessarily being completely polite about it. If you bought the barrel second hand, then you need to talk with whomever you got it from.
If you choose to continue to use the barrel - based on what you figure out - then you'll need to proof it again.
I'd probably start by putting the appropriate bottoming tap in the threads and see if it gets tight before running out of room, or if it stays easy to rotate all the way to the bore. If the latter, then I'd suspect the barrel has expanded a bit or it was a rather bad job of threading. IF the tap wobbles (loose or uneven tension in different directions), then I'd say you are done - the barrel is a failure.
Put a straight edge along each flat in the back and see if there is any evidence of expanding by looking at the side towards a light source. If you see evidence of bulging, even slight evidence, then you are probably done.
Sight the bore to see if there is any visual distortion in bore reflections of objects, or any evidence of lack of straightness. If so, then IMHO you are done.
Push a lead slug down the barrel from the muzzle and see if it gets looser as it gets to the threads. That may indicate an expanded bore unless it was deliberately made that way (some barrels have a deliberate taper or choke). Ask the manufacturer about the intended bore profile. Don't mention what you found out first. If they don't say choked or tapered bore, then you are done.
If there is any evidence of a problem with the barrel that isn't indicated to be a feature from the manufacturer, I'm afraid you are done. It isn't worth the risk.
If the barrel iteslf is provably ok at this point, then replace the breach plug. I wouldn't weld on the barrel or the breech plug, but that's just me. If welded, I'd be tempted to heat treat (anneal) to remove residual stresses.
Any chance there was an air pocket in the powder charge present when the proof test was made?
Gerald