The good news is from the EMPA-REG trial which showed that the chance of heart attack or stroke or renal impairment was significantly reduced by empagliflozin. These are useful outcomes, but also weight and blood pressure go down .
I understand your worries about the other things.
The risk of amputations was seen mainly in the canagliflozin study, but seems to be a slightly increased risk on any of this class of drug; but the risk is still small and the amputations are generally toes, not wanting to trivialise the problem. It is more common if one has hardening of arteries to feet so worth checking this; having said this, folk with hardening of arteris to feet are at increased risk of heart attack and stroke so might have more to gain from decreasing these risks with empagliflozin - swings and roundabouts. A french study looking at diuretic use (not at jardiance etc) showed that diuretic use was associated with amputations, so perhaps the problem works by decreasing the circulating blood volume ie less blodd in system to circulate to toes. So here, for me, the message is stick to the license ie do not mix with diuretics, and as always, one needs to keep an eye on feet and seek help at first sign of any bother.
The uro-genital infections are complex. Urinary tract infections are as common on these drugs as on placebo since poorly controlled diabetes is also a risk factor for UTI, but I don't know of an active comparison with another class of tablet. Thrush is a bother; this generally responds to cream eg canestan or daktarin, but can be really diffiuclt - this is rare, but at times the only resource is to stop the jardiance etc. Note if your GP wants to give you some fluconazole capsules for thrush (which are really effective) these do not mix well with statins. The risk of perineal necrotising fascititis AKA Fournier's gangrene AKA genitals turning black is overstated; the risk is low overall in the whole population and this risk is increased by being male, over 60 and by being diabetic; the extra risk from jardiance etc is minimal and just being diabetic is probably the greatest risk. But again, the message is that in diabetic people, if there is evidence of any infection anywhere, seek medical help ASAP.
It is important to highlight problems with drugs so that one can be informed, aware and look out for them, but often newspapers go for sensationalism and do not put things in perspective.
The DKA issue is twofold; firstly people with type 1 diabetes seem to get DKA on these drugs and it is not sure why, but this is another reason to stick to license. Thes econd problem is that one gets a slightly odd variety of DKA called euglycaemic DKA whereby the glucose levels are unremarkeable. So in folk on jardiance etc, if they are unwell, one needs to check the blood ketone levels. Overall, this is not a common problem and was not seen in the Empa-Reg trial. Although Empa-Reg and the Canvas studies were heart attack prevention studies, they give a lot of information about other problems of interest, and put things in perspective
Best wishes