I agree completely with mjc. Also, cyclopropane and halonium ions should have the same issue.
The epoxide (or any three membered ring) has an eclipsed structure automatically. So this should help explain what atoms are co-planar. The carbons are roughly tetrahedral, but distorted to make the strained ring.
This all relates to a thought process for determining the products of the opening of these rings, but I always found them kind of difficult to visualize. The nucleophile has to attack from the side of the half-chair that leads to an actual chair product, which has a trans diaxial arrangement of the new bonds. When drawing the two half-chairs, the nucleophile has a strong preference for attacking the side that sort of "looks" more sterically hindered, although this might be a superficial aspect of the picture.