My personal point on this: A 4.9k result is not a 5k race and thus not a 5k pr. If I want to run a pr, I only select races that are 'officially measured' or where I'm sure that the distance is really 5k.
While I absolutely see your point, for me that would leave a few races each year at most. The main thing I'm missing is seeing my actual shape compared to the prognosis in the graphs as well as seeing my yearly progression in the competition tab. It's not about an exact PR, but rather seeing my approximate shape in that year at a glance.
If a race is officially stated as 4.9k (or 5.1k), do you really think that distance is 'exact'? Measuring the distance by gps or with a bicycle will never be exact, so you don't even now if 4.9k is right or if it's only 4.8k. That's why I - personally - won't estimate a 5k pr from a 4.9k race.
I do know that you will never know the exact distance when the race is not measured according to the official guidelines. By collaborating the distance from GPS, rebuilding the course on a map and looking at the times of other participants it is still possible to get a best estimate. There is clearly some margin of error, but I think it is still good enough to get an impression of what your own performance was.