(a) Rashi on amud A states that the reason that the gizbor was not able to purchase the stones with money of hekdesh was because this would cause the seller to be moel. This is almost impossible to understand because what is taking place in the transaction is the transfer of kedusha from the money to the building materials which is what hekdesh is supposed to do with its hekdesh money.
(b) Furthermore, Rashi on amud B asks why the workers were not paid directly with the kodesh ktores. This suggestion seems to ignore the rule that Rashi developed on amud A forbidding direct purchases ( unless there is a distinction to be drawn between objects and services).
(c) Finally, Rashi's answer is puzzling. Even if there is not sufficient value in the ktores itself, the ktores could be given in addition to extra funds to make up the difference.
(a) Rashi's reasoning is indeed puzzling. I understood him to mean that the Gizbar would not purchase goods with Hekdesh money due to a Gezeirah d'Rabbanan , that the Gizbar may pay the money for the goods before the goods are handed over. In such a case the seller would be Mo'el, if he used the money before handing over the goods.
(b) Your distinction seems logical, especially according to the way I explained Rashi in (a). But this does not seem to be necessary, since Rashi is presently addressing the Hava Amina, that does not want to accept Shmuel's rule that "Bonim b'Kodesh v'Achar Kach Makdishim" (see Rashi).
(c) Olas Shlomo explains that since the Gizbarim would have to pay the workers part of their wages in any case, they avoided having to split the Chilul (or payment of the workers) into two by instituting the practice of paying the workers and afterwards being Mechalel the Ketores on the payment.