Hi Thomas,
I had to run this by the OptiSystem team because I was not completely sure! The answer is that the phase is modeled, as I said, as a phase noise which also means the instantaneous frequency is modeled by a probability distribution. This translates to the linewidth corresponds “roughly” to a FWHM of the spectrum. You can verify this with the OSA component in OptiSystem and the CW Laser source. Un-check the Resolution Bandwidth setting in the visualizer other wise you will get varying answers from the averaging.
To learn more about how we model the linewidth you can check out this commonly cited paper:
[1] C. Henry, “Theory of the linewidth of semiconductor lasers,” IEEE J. Quantum Electron., vol. 18, no. 2, pp. 259–264, Feb. 1982.