Despite the operating stress makes sense for mechanical engineering, most piping codes consider two separate causes of failure- collapse and fatigue; collapse is evaluated by sustained stress or sustained& occasional, fatigue is evaluated by calculating a stress range. Time validated this approach.
Where recognized, 'Restrained stress" is somehow an exception in such Codes because "fatigue", "sustain stress" or "sustain+ occasional" stress become meaningless- there isn't fatigue, pipeline is continuously sustained and seismic- for example- does not have inertial effects vs soil.
When you compare "unrestrained" operating stress with 0.9Sy this is not relevant in front of the Code that expects to address "collapse" and "fatigue". Why can be still calculated as operating stress in Caesar? I can only speculate it comes from very beginning development of software and they preferred to not touch it. True, in practice is rather confusing. As user, ignore it- that's all.