while you are correct that lab-leak is "theoretically" a distinct hypothesis, in this specific case the apparent splicing of pangolin virus parts together with bat virus parts is a highly suggestive smoking gun for man-made, especially when you consider that this lab does splicing experiments
From what I understand, this lab and those like it likely have/had a large backlog of samples they had received from field researchers but had not yet had an opportunity to identify, sequence, or otherwise mess around with. A procedural error in handling a sample in that backlog could have occurred, at least in theory.