You are looking at the apparent sun at sunrise, not the actual sun. Its projection upon the atmoplane.
The apparent sun at sunrise is on the rim of the sun's area of light and is racing along the equator or your latitude line to you. However straight your latitude line is in your local area where you can see will be how the sun appears.
If you were on the equator, and there was a race car (or jet ski) racing along the surface of the earth to you on the circular equator line, and you only see it until it is nearby, would you see it from the East or very near the East? If so, there is your answer.
So then how does the light 'bend'? The 'apparent sun' puts it in a different position (by apparently quite a large number of degrees) what mechanism brings it there? How does it bend it the correct way/amount depending on where you are? What evidence do you have that the sun we see at sunrise is not reflective of where the actual sun is?
Consider what happens when you walk into a dark movie theater. There is a projector at the back of the room, shining an image on the screen. When you look at the screen, you are not looking in the direction of the ultimate source of the light. The light from the projector doesn't need to bend at all in that scenario, just reflect.
You are looking at the apparent sun at sunrise, not the actual sun.
Where is the actual sun?
Probably further North.
Wise did have an interesting thread last year where not all of the shadows in daylight scenes were coming from the sun, and seemed to be coming from another direction. He had a bunch of examples of that happening. Maybe it's somewhere on the other FES website.
In the movie projector example above, there is still recurrent light coming from the projector source. If you look back at the projector, it shines light on you, despite you not being between the projector and the screen. It would be interesting if it could be determined which other direction the shadows were coming from in that thread, which I can't seem to find at the moment.