atom leveln called with too-low temperature for highest level, does nothing

[gary]

changes in r4204 announced when atom not solved by atom_leveln at atom_leveln.cpp:289. this produced vast amounts of output, which is now silenced by a debug enum posted in r4207.

this trip is caused by the highest level having no entry rates from ground. the routine exits without solving the atom. The total cooling is then zero.

This is a bug since there may still be significant cooling from low levels. The routine should decrease the number of levels until a highest level with a positive excitation rate is reached. This bug could affect the cooling, by either underestimating it, or causing noise as the highest level excitation rate was either positive or zero.

[gary]

fixed in r4214