It's generally not reckoned to be a good idea to use partial carb heat because this can potentially take the air temperature into a range where carb icing is more likely rather than less. Carb icing is most likely when the temperature drop in the carb takes the (humid) air to zero degrees C or very slightly below. If it is well above zero then it will not form ice and if it is well below zero then the ice is less likely to stick to the carb venturi. Full carb heat is nearly guaranteed to make the temperature well above zero in all conditions but partial heat can, in some circumstances, warm the air to the point that the ice is more likely to stick.
I've seen this asserted previously but I don't think the physics makes sense. Temperatures that are too low for ice are too low because there is not sufficient water vapour in the air for carb icing to be significant (in other words, the dewpoint is necessarily too low).
Heating the air, partially or fully, does not add water vapour. The only circumstance in which this might be an issue is when flying in glaciated cloud, where the air contains ice particles. If the ice particles are melted, they may refreeze in the induction system, hence it's better to let them go through as ice. But barring flight in glaciated cloud, I can't see how heating the air can do much harm.