What would be your recommendation on connecting the teensy to solderless breadboard?
You're dealing with so many unknowns, so many ways things can go wrong. So far at least damaged Teensy, maybe due to soldering, maybe from 5V power touching the wrong place, maybe something else. Use of known-problematic Traco switching power supplies, when used without additional capacitors. Now I2C devices not responding. But apparently no longer connected to the primary I2C port as in the prior photos and schematic. Why, who knows?
I see in your latest photo you've also trimmed away a substantial part of the Teensy 4.0 PCB material. That's quite risky, as it's a 6 layer PCB. There are 4 inner layers you can't see, where 2 of them are GND and 3.3V and the other 2 have lots of signal wires. That may or may not have an impact. The point is it's adding a lot more uncertainty!
The point of building a physically large solderless breadboard prototype is you can reduce the number of unknowns. You can check whether the hardware all works as designed, without the extra uncertainty of construction problems from soldering and aggressive modifications like carving away part of the PCB.
You can also get known-good software tested on the breadboard. For example, a month or two ago we had someone else report I2C wasn't working on Wire1. Turned out (after many messages begging for the full code to be shown) the problem was the Scanner example had not been properly changed, so it was using Wire1.begin() but then trying to actually use Wire somewhere else in the code. That may be your problem here. Or it could be those wires aren't connected, or 1 or both are shored to each other or something else, or maybe the hardware has been damaged some way during this construction. It's extremely difficult to follow where those wires go in the 1 photo of your latest attempt, and whether the Scanner program is correct is also blind guessing. With a huge number of unknowns, it's very hard to diagnose what's causing a project to not work. For those of us trying to help you over the internet, the harder it is for us to see what you've done, the less effectively we can try to help.
With a solderless breadboard, you could quickly & easily remove the wires connecting other I2C chips and check again until it works. If it's actually a software problem, you can quickly get to a test with only 1 brand new part connected and then focus your attention on the software side. Debugging is so much faster & easier with a solderless breadboard. Photos from a few different angles can also let us see how things are really connected and try to help you more.
As a general approach to experimental & prototype electronics building, getting each piece working and building up incrementally upon each prior success is a much better path. When something goes wrong, if you've added only a small amount, figuring out what has gone wrong is usually much easier.
Of course, if everything worked perfectly on the first try, there'd be no need to troubleshoot, no need to whittle down the huge number of unknowns. But that's clearly not the way things are going. So my recommendation it to consider changing your overall approach, because clearly this way of building is not giving you a path to success. Using a solderless breadboard as a first step may seem unnecessary, and if this were working well it would be. But unless you luck suddenly changes, I really believe you should consider using a different approach that lets you test as much as possible as you go and leverage each small success to chart a more reliable path toward ultimately making a successful project.