Well I'm glad it worked.
I've been doing PICs for a very long time, indeed my lineage goes back to the General Instrument days in the 70s. Inevitably over the years I have accumulated a modest but reasonable stock of bits and pieces for PICs of all shapes an sizes, from PIC10 to PIC32MZ EF. They all have their own quirks, and although there is some similarity or peripherals and functionality between devices, there are plenty of differences too. I've lost count of the number of PIC projects I've worked on, but it's well into triple figures, and there's no way my memory is good enough to remember all the differences.
Because of those differences, unless you happened to have fairly recent experience, it's unlikely that without the benefit of a data sheet you'd come up with a solution off the top of your head, however I did remember that for some peripherals on some devices like MSSP you need to set up TRIS correctly, but I couldn't remember exactly how.
To breadboard something simple like this up takes me about ten minutes: in fact, it took far longer to write the reply than it did to breadboard it up and figure out a fix, but that's just experience and having the right tools.
Back in the early/mid '80s when I first used I2C as a student, my sum total of test equipment amounted to a crappy 1k ohm/volt analogue multimeter, an LED+resistor and a home brew logic probe (a retriggerable monostable ~100ms pulse stretcher). With a healthy dose of imagination, you can do an awful lot with those simple tools, but debugging I2C in any depth, not so much. Back then, we didn't have the luxury of on-chip in circuit debugging on an MCU either, that was still about fifteen years away, but you could spend an arm an a leg on an external hardware in-circuit emulator.
Nowadays in real terms, you can pick up a reasonable 2nd hand CRO for the same as I paid for my multimeter!
While it's possible to debug your problem with very rudimentary tools and a reasonable amount of creativity, in view of the almost obscene cheapness of some tools available nowadays where the retail cost is frequently less the the BOM cost, you'd be nuts to want to beat yourself up. As a caveat to that I would say that doing it the hard way you will inevitably learn a lot more, mostly by gaining a far deeper and granular understanding, as well as dramatically improving your troubleshooting and analytical skills.
Again, thank you for your kind comments, but it wasn't much effort: I like to take these opportunities to keep my own skills polished.