it is also the price of what is currently for sale, most shares of most companies are not actively for sale at their current price.

Every market participant has a buy price and a sell price for one share influenced by their personal opinions and the state of the whole market; each at a different level and trading price is only calculated on actual trades. So all the shareholders that are unwilling to sell for any of the current offers do not influence it.

Yet if you were to buy 100% of the shares then you would eventually have to climb up to their price.

This is an interesting point I never really thought about. There is a distribution of buy prices and another distribution of sell prices. But if we plot these on the price axis there will be NO overlap because all shares in that region have been traded. I wonder if there's a way to sample these distributions to get the bigger picture.