Digital color works by separating light into three channels: red, green, and blue. This roughly mimics the way the human vision system creates the perception of color. Our vision system compensates inherently for different-colored light sources using environmental cues, but when you look at a digital or printed photograph, those cues aren't there, so we see a color cast.

"White balance" compensates for this by pushing up or pulling down the values in some of the channels in order to mimic the effect of a more neutral light source. For example, if the scene in actuality is very blue, in the rendered image, blue is suppressed and red and green (which together make yellow/orange) are increased.

At the extreme ends of the scale, the recorded light is going to be almost all in one channel — in your case, all red. There just won't be very much blue or green data available. Pushing that up to normalize the white balance will at best add huge amounts of noise. (That's what happens when you amplify a small amount of signal.)

The precise limit of a given app or camera is probably somewhat arbitrary, but there's a good reason for a limit, and the available range generally represents the range of reasonable adjustment.

In cases like this, if you can't accept the color cast, it's often best to just convert to black and white.