With NiCd cells, where the energy supply is > to >> what is needed during the charging cycle AND battery temperature can be kept reasonably low, solar charging of NimH is similar to charging NimH in any other environment. In such cases, if people tend to not use a "proposed "new" method it's probably because it's not as good as alternatives.
In the case of Nimh, they are usually charged at constant current (CC) until certain conditions are met. They exhibit a declining current drain only if charged at CV and this is not usually what is done because CC with defined termination is better.
In the case of small solar systems where the charging power (= instantaneous rate at which energy can be supplied) is variable and sometimes (or always) less than the battery might safely be charged at and/or where battery temperature may be high and sun affected, then normal charging termination methods are not suitable. eg negative delta V fails when it may be caused by end of charge OR clouds. Delta temperature fails when it may be caused by end of charge or sun. Absolute temperature likewise.
"Coulomb counting" requires more complexity and may be only somewhat viable in solar applications. Where battery temperature is high when charging (eg if device has integral solar panel and is placed in sun to charge) and then discharged under cooler conditions (as in a light which is charged by day and used at night) then as temperature rises during charging, charge efficiency falls and if you used simple coulomb counting then you'd experience severe to very severe undercharging. You could temperature adjust for probable charge efficiency, but you are then in such uncertain territory that the simple method below will usually work as well or better.
In such cases the best (if not marvellous) end of charge indicator is probably battery terminal voltage under charge, modified to account for temperature and charge rate. I use 1.45 V/cell at charge rates in the order of C/5 - C/10 and dynamically modify that somewhat to account for charge rate as this voltage is approached under charge. This allows a degree of variation between different manufacturers products to be accommodated. What is fatally bad is to set a voltage which is so high that it is never reached and on a really sun filled day with partially charged batteries initially the batteries "cook" on overcharge for much of the day. Temperatures in portable solar charged devices exposed to full sunlight can easily exceed the temperatures at which NimH will usually tolerate. lights You can buy high temperature cells but the best solution is a design which maintains lower battery temperatures by whatever means. Fo AA NimH cells under about 1600-1800 mA/cell a degree of overcharging or trickle charging is tolerable as means is provided of recombining gases liberated when end of charge is reached*. Modern high capacity AA NimH cells over about 1800mAh (and 2500 mAh+ is now common) do not have the recombination systems (in order to allow more room for active chemicals) and NO overcharging or trickle charging should be allowed. If the cell "gases" the gas is NOT "recombined", internal pressure rises, the cell vents (or explodes) and electrolyte is irrevocably lost.
I designed these solar powered portable lights and after much experimentation concluded that the battery terminal voltage method was most liable to best meet end of charge termination requirements. Whether this would apply to other systems depends on the factors discussed above.