I had similar experience with DFI k6bv3+/66 MB. I got it in horrible shape, all stripped out of elec. caps and inductors, after repopulating missing parts it booted, but only with L2 disabled, couldn't find any cut traces on PCB, swapping ram/cpus didn't help, reducing clock freqs didn't help either, voltages were in range at ICs supply pins. Didn't figure out what was the cause of non booting with L2 cache on, because it mysteriously died when I was replacing TAG chip 😕

I would have tested one more thing if it wouldn't have died. The first time I powered it on, it had two pins shorted on super I/O controller, which prevented booting PC from any media, separated them and then I could boot PC only with L2 disabled. I'd try on it now to completely disable integrated I/O and install external I/O card just to see the result (I know that it doesn't make much sense when it concerns L2 cache, but who knows at this point).

Nice work there, congrats! What do you use for re-soldering those chips, hot air or just a soldering iron?
As for your problem, i guess a bad via. Have you checked the traces that pass under the capacitors? I have a SS7 board (forgot which one!) that has some traces corroded by a capacitor near the AGP slot...

SuperIO chip is the last major part that I haven't yet replaced or otherwise tried to debug, won't hurt to try but hopes are low haha.

I desolder the parts with a heatgun and put them back with a regular iron. I have found heatgun is useless for putting stuff back unless you have solder paste and stencil. Drag method works very well.
No capacitors have leaked there, but I will go over all the vias soon. Not looking forward to it 🤣