Why doesn't planet Earth expand if I accelerate upwards when standing on its surface?

Spacetime curvature makes this possible. Here's an analogy. There are two paths on opposite sides of the equator, at a constant distance from it. Someone walking east along the path north of the equator will have to continually turn slightly left to stay on the path. (If that isn't obvious, imagine it's so far north that it visibly circles the pole.) Likewise, someone walking east on the path south of the equator will have to turn right. Two people walking side by side along the paths will stay the same distance apart, even though they're constantly turning away from each other. This wouldn't be possible on the Euclidean plane, but it's possible on a curved surface. That's what happens in general relativity, but the direction they're walking is the time direction, and the turning is acceleration.

According to General Relativity I am being accelerated upwards by planet earth while writing this question.

According to general relativity you are being accelerated upward by the normal force. This is exactly what happens in Newtonian mechanics.

One difference between the two is that Newtonian mechanics deems gravitation to be a real force while general relativity does not. A frame based on a person standing still on the surface of a non-rotating rogue planet is very close to being an inertial frame in Newtonian mechanics. The person is standing still because the upward normal force and the downward gravitational force cancel one another.

An inertial frame in general relativity is comoving with a stream of falling apples. A person standing still is accelerating upward from the perspective of a stream of falling apples. This upward acceleration must necessarily be the result of a real force, which is the normal force.

But a curious person on the the other side of the planet relative to me would have the same experience. That means we are accelerated in opposite directions, although earths diameter do not seem to increase. How can this be?

Another key difference between Newtonian mechanics and general relativity is that inertial reference frames span the universe in Newtonian mechanics but are local in general relativity. Mathematically, "local" means infinitesimally small. The concept is a bit more expansive in physics, where it means small enough that instruments cannot detect accelerations due to differential gravity (e.g., tidal effects).

Nowadays, Einstein's elevator car thought experiment doesn't quite cut it as instruments capable of detecting the differential gravity across an an object the size of an elevator car have been developed; this was the basis of the European Space Agency's Gravity field and Ocean Circulation Explorer (GOCE) satellite. A relativistic inertial frame with its origin at a person's center of mass standing still on a planet does not extend to a person standing still on the other side of the planet.