Based on your description:
- it "passes signal"
- you get measurements indicating a functioning LFO
Note that the "bypass" is essentially a mechanical version of how many pedals electronically produce bypass/engage. That is, instead of using a JFET to block/enable the wet signal, a switch is used to either connect the wet path up to a passive mixing node...or not. So, even if you had flawless modulation, if the footswitch - which is older than the majority of members here - was not making good reliable contact, you would get "signal passing" in both bypass and engage mode but not any audible phasing in effect mode.
It's an easy test to carry out. Just bridge the footswitch and see if that restores the effect. If not, then the next culprit in line would be a bad 741. The test for that is also easy: stick your probe at pin 6 of each of those 741s and see if you get signal at its output.
Given the two things it
does do, my money is on one of those two issues. Yes, there could be an issue with one of the 2N4303 JFETs, but the odds of all of them crapping out are very low, in which case you'd still get some audible phasing, just not as strong. So it's gotta be something creating a blockage of the wet signal. It
could be the 150nf cap before the footswitch, but I doubt it. Still, put that the third item on the list.