The coning is actually a measure of many factors, as you have carefully stated, AnFI, but in a case like the Greek Apache, it probably indicates that the power demand that the pilot made was excessive (and yet not enough) so that the engine power (torque) limiters allowed the rpm to droop, thus making coning angle higher because the centrifugal stiffening was much lower as the rpm was pulled down.
All this is great speculation, the data recorders on the aircraft should have captured it all and made it available.
You raise interesting points, but a central one is mostly overblown and should be discussed.
Rotor stall does not affect the large percentage of CFIT accidents, most rotors stall at load factors far above the load factor that the engine power can produce. Unless the airspeed is very much higher than Vy, the speed decay that can fuel the "extra" load factor is just not available.
Here is a chart that shows the energy available for "fueling" maneuvers from the typical sources. Note that below 80 knots, kinetic energy is vastly out weighed by engine power and even by stored rotor energy.
Here is actual maneuver data from a Utility helicopter during air combat trials, plotted against Ct/Sigma which is actually a stall factor. It represents the absolute maximums that can be squeezed from the machine by an experienced test pilot. The rotor stalls at about .18 to .21 Ct/Sigma, so the typical low speed maneuver maximum is far below stall. Note how the shape of the typical maneuver is the same as the engine power shape in the previous chart? No coincidence! For comparison, .15Ct/Sigma is about 2 G's in this case. BTW, advance ratio is the tip speed ratio, so that .2 is about 85 knots in this case. Note that when speed is higher than Vy extra G is available as transient, powered by a deceleration.
