Reason why $f(x)=x^2$ is not uniformly continuous. Consider this calculation: for any $x\gt 0$ define $\bar x=x+\frac1x$. Then \[ \bar x - x = \frac 1x, \] and \[ f(\bar x) - f(x) = \Bigl(x+\frac1x\Bigr)^2 - x^2 = 2 + \frac1{x^2} \gt 2. \] If $f$ were uniformly continuous, then we could choose $\varepsilon=1$, and there would be a $\delta\gt 0$ such that $|f(\bar x)- f(x)|\lt 1$ for all $x, \bar x$ with $|\bar x-x|\lt \delta$. But our calculation above shows that if we choose $x$ large enough (i.e. $x\gt \frac1\delta$), then we have $\bar x-x\lt \delta$, and $f(\bar x)-f(x)\gt 2\gt\varepsilon$.
If $n, m\ge N$ then we have $x_n, x_m\in(0, \delta)$, and thus $|x_n-x_m|\lt \delta$. This then implies $|f(x_n)-f(x_m)|\lt \varepsilon$ for all $n, m\ge N$. Since this argument applies for any $\varepsilon\gt0$, we have shown that the sequence of real numbers $f(x_n)$ is a Cauchy sequence. It follows that $\lim_{n\to\infty}f(x_n)$ exists.
Consider the function $g(x) = f(x) - P(x)$. This function vanishes at $a,b,c,d$, so by Rolle’s theorem there are $p,q,r$ with \[ a\lt p\lt b\lt q \lt c \lt r\lt d \] such that $g'(p)=g'(q)=g'(r)=0$. Apply Rolle’s theorem again, but now to $g'$, and you find that there are $s\in(p,q)$ and $t\in(q,r)$ with $g''(s) = g''(t) = 0$. Applying Rolle one more time then gives you a $u\in(s,t)$ with $g'''(u)=0$.
This implies $f'''(u) = P'''(u)$. But $P(x)$ is a quadratic function, so its third derivative vanishes. Therefore $f'''(u) = 0$.
The interval $[a,d]$ is compact, and the function $f$ is continuous, so it attains its minimum at some $x\in[a,d]$. Since $f(b)\lt f(a)$ the minimum is not attained at $a$, so $x\in(a, d]$. Similarly, $f(c) \lt f(d)$, so the minimum is also not attained at $x=d$. Therefore $x\in(a,d)$ so the point $x$ at which the minimum is attained is an interior point, and we conclude that $f'(x)=0$.