Anything less than 1.0 is outright unstable, if the calculator is accurate. I see now that Chuck used the Oehler Ballistic Explorer. No idea why it thought the moly would be different. The only thing I can think is that it automatically reduced velocity, as moly bullets tend to run about 50 fps slower if you don't adjust the load. But if you do adjust the load to get the same MV, it should make no difference.
The 20" barrel the OP has should produce Jimro's 2750 fps with a near full load of Varget or anything else that passes about 2890 fps in a 24" standard velocity barrel, but the starting load will be about 2630 fps. The Miller estimator gives that a GS of 1.2, which is not unstable, but not good. A Sierra tech once told me they want to see 1.3–3.0 for hunting accuracy, and 1.4–1.7 for target accuracy.
Geoffrey Kolbe's more detailed twist estimator, based on the late Robert L. McCoy's McGyro program, thinks the 2155 will actually be a little more stable with a 14" twist all the way down to about 2050 fps, where it puts the GS factor at 1.3. It thinks 1.4 would be all the way up at about 3300 fps. This is based on entering the bullet's full set of dimensions from Bryan Litz's data. Note that I had to use a density of 10.0 to get the right weight estimate from those dimensions, as found
using Kolbe's drag calculator.
Kolbe's calculator also finds the 2156 unsuitable. Note that this bullet has a lot of empty space in the long nose, so a density of 9.7 is necessary to get the weight right if you try it in Kolbe's calculator.
Litz's measured lengths are a little longer than the ones Chuck measured, and this accounts for part of the difference. He has the 2155 at 1.131" and the 2156 at 1.210".