It sounds that what you're really debating is the actual definition of underground infrastructure, perhaps what steve is adding to the next version is actually "low gravity" infrastructure, and people would like a third type of "underground infrastructure" which could be added to any type of planet regardless of gravity cost which can be used for protection from planetary bombardment, lowered thermal emissions, and fixed colony cost regardless of other factors.
Something like this might be useful for putting people onto venus without orbital habitats or large scale terraforming, but would make putting population onto already colonisable worlds possibly far to easy. Whereas the proposed underground infrastructure is basically adding to the game something new, allowing previously uncolonisable worlds available. Besides extremely high gravity worlds would basically be completely uninhabitable without some kind of amazing antigravity technology, whereas very low gravity environments are already inhabitable, at least for the short term.
Edit: I've reading about the energy requirements for speeding up a planets rotation, well the moon actually, it seems 10^25 joules would be required merely to speed the moon upto 24 hour rotation. 10^28 jules would actually deorbit it or possibly be enough to actually blow it into chunks which would basically deorbit anyway. The energy needed to speed it's rotation enough for artificial gravity would be far more than the energy needed to merely push it into an orbit which would make the surface more habitable, at least from a temperature standpoint. Perhaps a new terraforming option could be used for very small low gravity bodies, which allows changing the rotation period enough to allow artificial gravity?
I'm not debating the utility of the proposed underground infrastructure, just the science behind them, it would be cheaper to merely put your colonists inside giant centerfuges if the body is too large to actually move.
Edit 2: Some very basic and possibly flawed math leads me to the following conclusion.
In the case of triton with a gravity of .08 being just barely below standard habitability has a mass 29% of the moons, and would require perhaps 8.5% the energy to rotate.
For a body like ceres with .03 gravity and a 61.5th the mass of the moon something like 4 millionth the energy would be required. (unless my knowledge of physics is horribly wrong) leaving something like 2.5^19 jules.
Tsar bomba was estimated to have released 4.2^18 if it hadn't had it's yield halved by the addition of a lead tamper, so conceivably it's within reason that a body that small could actually be given enough spin for artificial gravity.