The commonly used Langelier Santuration Index, the less commonly used Ryznar Index, and our own CSI that we use here are great at predicting the scaling propensities of water, not so great at predicting the aggesiveness of water against plaster surfaces, and totally useless to predict the corrosiveness of water on metal surfaces. However, since this is actually a very complex subject our industry, which doesn't always have the brightest bulbs working in the field, tend to incorrectly simplify things and then you get the confusion about calcium protecting heat exchangers from corrosion. It is actually the deposit of scale that will form because of the elevated temperature in the exchanger that can help prevent against corrosion but it's pH and the absence or presence of other ionic species in the water (such as sulfates or chlorides) that are much more important factors in how corrosive toward metal the water is.
As far as vinyl liners, CH doesn't matter at all (except at the high end where it can lead to scaling conditions) but low pH will damage a liner by leaching the plasticizers out of the vinyl and actually attacking the vinyl itself.
There does seem to be evidence that CH is important in fiberglass pools to help prevent cobalt spotting and to reduce the propensity of fiberglass to stain.