Look. At some point it has to happen. If you have a dino that has flat teeth and all its ancestors had pointy teeth, then at some point in time, a baby dino was born with flat teeth.
We don't see piles of bones with dinos have less pointy and more flat teeth in a gradual change over eons, we see pointy teethed dinos, then, woah, two layers up we see the same dinos with flat teeth.
This is what people mean when they say the critters appear fully formed. The minute changes over time you keep mentioning don't exist.
We don't see that anywhere.
This is why puncutated equilibrium is back in vogue.
And so, yes, at some point we should see a salamander born who has a major morphological shift from it's parents (an exoskeleton for example).
You can't just keep saying minute changes too small to see that eventually lead to a major morphological change. You are simply rephrasing Zeno's Paradox, and that was solved a long time ago through the discovery of limits in mathematics.
And what percentage of dinosaurs that lived and have been found as fossils.
This is not a cheap evasion. How many fossil dinosaurs have been found, and how many dinosaurs lived?
When you do find fossils, you tend to find a lot. Chances are they were preserved at almost the same time by some flukey conditions. Oddly enough, they all seem to belong to the same era.
You fail to mention two things. Fossils never appear out of their expected sequence, and new ones are found every year that fill expected gaps.