LTSpice Simulation LT8365

we are going to use the LT8365 as a Step-Up Converter for generating dynamic Voltage Supply up to +/-110V. In simulation with LTSpice, we got two issues, which we like to address here:

To filter out the Switching Noise and the resulting Voltage Peaks at the output of the LT8365, we use Pi Filters. It works well assuming ideal components in the simulation. But if we use Spice Models of real components by the manufacturers, a damped, but significant oscillation of about 3 Ghz remains. This is, beside the real component behaviour, caused by the Switching Speed of the LT8365 of about 300ps. Is this Switching Speed similar to the real Switching Speed of the LT8365 or is it just approximated by an "ideal" Switch, so that in practice this oscillation will not appear?

The second issue is, that with reducing the Output Voltage by influencing the Feedback Voltage with a Resistor Network, the Output Pulses begin to jitter and a oscillation between the switching pulses begins to appear. This only happens in simulation, when the coil current drops down to zero. Is this a known behaviour of the LT8365?

Thank you for your response!