Face mounting the plate to the side of the rail (photo #1) using only one face's T slot is one of the fastest ways to just about gaurentee that you have problems with the T-nuts ripping out the edge of the rail. I'm speaking from experience here; the screws and T-nuts are a lot stronger than the rail, and will (eventually) pound their way out when dynamically loaded in that way.
You need to engage at least 1 other edge (the bottom); adding the top as well would be great. You need to remember; when using the T-nut connectors, you are only achieving a clamp type connection, NOT a thru bolted connection.
Well, I was suspecting it'll be the weak point. But I have to constantly remind myself to not over-engineer. Chinese kids cars at wallmart use all plastic, and hold up just fine (although at slower speeds).
What if bolt it all the way through in two places with 1.5x1.5 backing plates on the other side?
Engaging more than one face of t-slot means I can't use a flat plate anymore. That complicates things a lot
As for your spindle idea, two changes for strength; make the current 1" thickness at least 1.5", and drill the pivot hole straight thru. If you leave it at 1", then you'll only have 3/16" (.1875", ~.47mm) per side; that's way thinner than I'm personally comfortable with. This piece needs to be extruded matl.; most grades of cast won't be strong enough.
The two tapped holes? Built in stress points, and trying to keep them tight will be problematic. One smooth thru hole 1/2" (.500", ~12.7mm) will be considerably stronger, as well as much easier to keep tight, as it would be one steel thru bolt (preferably a grade 8). Again, adding the extra 1/2" takes the thickness per side from 1/4" to 1/2" (6.35mm up to 12.7mm).
The kingpin whole(s) is 3/8 or 7/16, that leaves plenty of wall, over 1/4" in each case. Multiplied by 2" of height - i don't think it's going anywhere. But it's possible to go with 1.25 thick plate, I guess.
I was considering two holes vs straight through kingpin and decided two will be stronger - as they leave a solid crossection in the middle of the part. And this is all static connection - the rotation will occur at the balls in the rod ends. But I can be wrong. Straight through is surely easier to do.
For the axle hole - yeah, 1.25 aluminum will be probably better.
And it'll make the business of adding control arm easier. I'll just drill two small holes on 1.25 side and attach the arm.
I want arm to be separate and replaceable part. First of all - it WILL bend during accidents. And second it'll make steering setup/adjustments easy.
And the final thought? If you insist on CNC'ing that part, that's your call. If you're more interested in just the "Fit, Form, & Function" rule, I've got the equipment to do the major work; how you decide to polish them would be your call.....

Pat
I don't insist on anything. I just need to make this part. The way I simplified the part - it can be made from flat piece, with simple tools you have! I just need to buy rims first to finalize the dimensions. Right now i only have tires.
Do you think it's necessary to drill the axle hole through? I was hoping it can be just taped hole, so I don't need no nuts, just one bolt and lots of locktite. The wheel will have bearings, so this is non-moving part will only need to withstand vibrations.