One thing to note: that hasn't always been the case. This is something that can change.
It really started in the late 1970s with the Friedman Doctrine.
The Friedman doctrine, also called shareholder theory, is a normative theory of business ethics advanced by economist Milton Friedman which holds that the social responsibility of business is to increase its profits. This shareholder primacy approach views shareholders as the economic engine of the organization and the only group to which the firm is socially responsible. As such, the goal of the firm is to increase its profits and maximize returns to shareholders.
I'm trying to find the story I listened to about this on NPR a few years ago, but it essentially discussed how this doctrine was taken up after the stagflation in the 1970s (particularly as Reagan was heavily influenced by Milton Friedman). The main point was that it seemed like the traditional economic system was collapsing at that time, and Friedman's ideas argued that it was because businesses were not focused enough on profits. Instead, many businesses were trying to be part of a broader community and work on doing things that were good for the public. Friedman's idea was that this was too economically inefficient and that a businesses only ethical obligation should be to make money for the shareholders, and that the shareholders could decide for themselves on how to help the public.
This went over very well with business leaders, and it helped ushered in the Gordon Gecko era of unironic "greed is good".
Fahrenheit is also based on water's phase changes, but the 0-100 range just falls nicely around the range of common ambient temperatures. The basis in water is nice because it's abundant and thus makes calibration of a thermometer easy. My contention is merely that the specific values of the phase changes are not so important that it makes the Celsius scale inherently better. I like that the ambient temperatures outside fall nicely throughout the 0-100 range in Fahrenheit, and I think that is just as valid an argument as water being liquid within this range.
And perhaps I'm particularly swayed by this argument because I live in a place that has cold winters and hot summers, so I see the full range of 0 to 100 in the weather. I'm also not going to pretend that growing up using Fahrenheit is not the main reason for my continuing usage of it.
I just wanted to point out that I'm convinced by the arguments in favor of the metric system for everything except Celsius. For that one, I just don't think water is as compelling an argument as is always presented.