I remember Harold Young (ex NSWGR CME) talking about loco designs. While he is famous for the 38 class he considered the 58 class was a much more difficult design. In fact he stated that the 58 was closer to a 38 than a 57 class.
One the problems with the 58's was that the crews would try to drive them like a 57 with partially opened regulator and long cutoffs. When they should have been using the 38 class technique of full regulator and short cutoff and therefore taking advantage of the expansive power of the steam. This was very hard for goods train crews who had not had experience with this technique.
Furthermore, by the time they were put into service, dieselisation was in progress and no monies were allocated to fixing up the minor problems with them or crew training.
The 58 class was doomed from the very beginning. It was a difficult design because they started with 25 underframes designed for the 57 class and tried to build a locomotive to fit tighter clearances. The NSWGR basically lied about their intentions and received 50 cast frames from the USA during the war. 25 were for 38s, and proved useful although not until after the war. The 25 57 class frames were delivered and the NSWGR realised that it had to build something with them, although clearly they didn't need 25 more 57 class.
Young clearly believed that by incorporating various changes he could compensate for the much smaller cylinders, but I think he was wrong in this belief. In particular he expected more from "Chapelon" style improvements than he got, possibly because his detail designers hadn't taken account of the different steam conditions - Chapelon's work was largely based on improving the low pressure side of compound locomotives, and the 58 was of course a simple.
To overcome the basic difficulty of the Gresley version of the Holcroft valve gear, it should be located behind the cylinders so that the centre valve motion was not affected by lengthening of the valve stems due to heating. The 57 had its main brake cylinders behind the cylinder block and of course, the 57 frame didn't allow an easy redesign.
The 58 class were not abandoned because of the introduction of diesels. The 58 class construction was stopped well before the 59 and 60 classes were delivered. It was clear by 1951 that they were a mistake.
Some of the design work was very clever, but the rack and pinion valve gear did not overcome the problems of the original Holcroft gear. The operating insistence on giving 58 class the same loads as 57 class and blaming any shortfall on the crews was probably the worst thing they could do.
As late as 1955, C.A. Cardew was asking for them to be rebuilt as standard 57 class, but the delivery of the 42 class that year showed what the real answer would be.
I don't believe a useful locomotive could have been built with the 58 design parameters. The frames fitted with two cylinders and a modified 38 class boiler with a longer combustion chamber might have resulted in a useful high speed freight and heavy passenger loco for the south that could have run on the north after the bridges were replaced.
But when the question is wrong, the answer doesn't matter.