from what my understanding, UV spectrum is absorbed by bonds.
metallic ions (and ions in general) a full outershell, making it very stable and have little absorbing potential. metals in AAS however goes through the burner and are reduced to ground state, hence why AAS is very useful for detecting metallic ions.
I'm presuming that UV spectrum are not absorbed by ions. however, transition metals such as copper form colourful complexes, and the bondings in these complexes absorb UV, so *some* metallic ions can be analysed by UV
for q3:
a) since drugs are not metallic ions, AAS would not be used. but as for identification, i thought IR would be more suitable, as it can be used for identification, UV-vis tells the concentration, not molecular structure
b) C13 NMR could be used, but IR is easier, more cost effective
c) metallic ion = AAS. if it was permanganate (or other compounds), UV-vis would be suitable.