Thanks for the screen shots. It's clear that the IFR is advising you that for the given sweep time, it suggests that the RBW needs to be increased for accurate results. I don't see any such indication on the Rigol. So, it is either ignoring the sweep time you asked for, or it is giving you that sweep time and the display update rate is being driven by how fast it can reset the LOs for the next sweep (which may be on the order of 10-100ms). I wonder if you can visually see a difference between a 100ms sweep time vs. a 10ms sweep time, given how fast it updates the LCD screen... It appears really to be a question for Rigol tech support to understand what it is really doing. I would expect that 10ms sweep would be too fast for a 100kHz RBW.
On on older HP swept SA that I have here, using a 83MHz span and 100kHz RBW, the auto-coupled sweep time is just over 20ms. If I manually adjust it to be faster, it lets me do it, but also gives the warning that the measurements are uncalibrated.