Scott:
I don't disagree that a reduced voltage starter whether via resistors, SCR's, autotransformer, wye/delta, two speed starter, one arm bandit, etc. is a good idea to reduce starting currents.....
And your right, you can install resistors in series with the output of the inverter, along with a switch or relay in parallel to short them out once the motor is up to speed. Or better yet, place an AC current meter in series with the output, and close the shorting switch once the current has settled, and the motor is up to speed. To me this sounds labor intensive, and I would worry about forgetting to close the shorting switch (I'm getting old :-). If the shorting switch was not closed, overheating of the resistors and potential fire is only one of many possible scenarios.....scary thoughts.
However, if you are familiar with failure mode of inverters, their current sensing circuits, and fault trip circuitry, you will realize that to say inductors are not a good choice is not entirely correct. IMHO
An inductor by definition "resists a rapid change in current". It is this rapid change in current (Di/Dt) that can cause damage to the output transistors, especially when switching into a low impedance load (stopped motor for example).
By placing an inductor in series with the output of the inverter, we are reducing the rate of rise of this high current spike each time the output transistors are "fired on". (I'm talking about milliseconds and faster speeds).
This is easier on the output transistors, and slows & limits the current spike to allow the current sensing and fault trip sensing circuit time to react before damage usually occurs. Picture a line graph where the amps rate of rise was near vertical, and after an inductor it becomes sloped, more 45 degreeish.
In my limited single phase inverter experience, it appears the more expensive the inverter, the more these output "filters" are included. What's to say we can't add them on a cheap inverter? :-)
An additional benefit to placing an inductor on the output of a MSW inverter is the effect that it will "clean up" the output waveform some. At only 3-4% impedance, you will not achieve a clean sine wave, but an induction motor will be much happier with the improved waveform.
In conclusion, I agree a 3-4% impedance inductor will not reduce the high motor starting currents from 700% down to 100%, yet, an inductor will help to reduce the damaging peak currents that kill power transistors commonly.
As I have seen it said here before "your mileage may vary".