Alex,

What you need to know is the following:
You cannot have VNAV engaged at all if your performance data is incomplete and if you pay attention to the FMAs you will see clearly that the VNAV is (in green) ARMED.
That altitude missing that you correctly noticed is code related that is not computing correct to a wpt manually created as a PBD.
That 8K FT that I'm talking about is a "hard altitude" for a wpt that was busted, again a coding problem to be easily fixed.
We both agree that LNAV needs pampering.
As a real pilot, I assure you that with the latest versions of the FMC that issue not being able to comply with 2500B in our case is not "perfectly fine/normal and if you look at the ATP/L standards it's not. On top of that if the FMS is not capable to make a crossing restriction you will get UNABLE NEXT ALT.
As a real pilot, jets in real-life comply indeed with crossing restrictions. I have to admit that very small deviations are encountered under turbulent wx conditions.
As a real pilot, I would like to call to your attention that there is no "VNAV path-vector", we call it vnav deviation scale with a vnav pointer.
The bottom line here is that we're looking to have some items fixed for a better sim experience.