I agree that not all mutations have to be defintely self sustaining at first, but for such rare and deleterious mutations such as above, the frequency of the mutation will just not propagate into the population because the mother does not have a high progeny output. The mutation can arise again in the population, but it will not become anything if natural selection favours so strongly against it. So I see what you're saying in that it could change from being fitness reducing into being sustainable, but only if the mutation wasn't so harmful that it ends the production of progeny after few generations.
It's just hard for me to believe that, in birds where egg laying is such a specialized mechanism, that such a dramatic and fitness reducing incident could be the beginning of internal gestation borne from oviparity, especially if the mechanism behind it is deleterious. It just makes more sense to me that smaller, less harmful mutations accumulated over a large period of time in the right environmental/biotic situations allowed internal gestation to be favoured and thus evolved.
But it would be cool if this incident was largely genetic. Sometime in the future we'll be able to accurately relate sequenced genomics to complicated devlopmental genetics such as these.
.... And that may be the time where we find out which genes cause people to be gay too : D