The trick with Clip Fix seems to be using Amplify afterwards to bring-down the “restored” peaks so they don’t clip again when you export.
Clip fix probably won’t fix all of the audible distortion but it may help.
although there are still a few clipped points in the latter part of the recording.
If you can’t hear the distortion, it may be best to leave those alone. You can try to see if it helps…
I recommend you split the recording into songs, or at least logical sections. At a minimum, I’d make separate files for the 1st two loud songs, the 3rd song with the volume-change in the middle, and the last part which is mostly OK.
Work on those sections separately to fix the clipping and to fix the volume change in the middle of the song. At this point, don’t try to match the volumes, just try to make each section sound it’s best.
How can I un-clip the whole performance and normalise the volume so that all songs appear to have been recorded at the same level?
Don’t worry about how they “appear”.
Once the clipping is fixed as much as possible, and that 3rd song is evened-out, run the Amplify effect to normalize/maximize all of the songs/sections.
Then if they don’t sound equally loud, choose the quietest-sounding song as your reference and adjust the louder songs down to match (by ear).
Then re-join the sections with cross-fades (cross-fading the crowd noise/applause between songs). That will allow you to remove any excessive gaps or excessive talking etc.
Sometimes a long crossfade (several seconds) works best in a situation like this. And sometimes, I’ll steal some applause from one part of the performance and re-use and/or blend it in somewhere else.