My experience would suggest the opposite as heating the assembly with the oil already applied to all the joints results in a steady stream of bubbles from the threaded interface. Also when apart you can usually see that the oil has penetrated the threads to some extent.
Also, rightly or wrongly I'd expect the volume displaced by the air as it's heated to be greater than that of the metal components, if that makes sense.
It doesn't work as you think it does.
The reason a pedal is stuck so fast is because of friction welding. That means, the threads on the steel pedal are fused to the threads on the aluminium crank. It is caused by fretting, which in turn is caused by the movable joint. A pedal should have had a taper on the end like a wheel nut. Instead it has a square shoulder and this shoulder moves significantly during pedaling. It (rouge and gouging) can be seen when you remove the pedal.
From the fretting, the threads weld. However, they only weld on the pulling side of the thread's "V" The other side essentially doesn't even touch. Fluid may well penetrate here but it won't contribute to loosening because it can't penetrate the welded sections, which are the problem after all.
When the crank is heated, the expansion does two things. It enlarges the hole and it thickens the crank (i.e. lengthen the hole). This breaks the weld.
Millions of liters of useless "penetrant oils" are sprayed on components each and every day, yet very little of it does anything other than to stink the place out and make you feel like you're making a positive contribution to solving the problem.
