The limit isn't too bad using l'hospital's rule, but I was wondering if there was a way to do it without l'hospital's.
Looking around the section limits without lhopital's, it seems usually evaluating without requires some clever factoring, while here the $\arctan$ seems to muck things up.
Here is the evaluation using l'hospital's in case someone visits with this question: $$L=\lim_{x\rightarrow 0}(1+\arctan(\frac{x}{2}))^{\frac{2}{x}}\Rightarrow \log(L)=\lim_{x\rightarrow 0}\frac{2}{x}\log(1+\arctan(\frac{x}{2}))\\ \Rightarrow \log(L)=2\lim_{x\rightarrow 0}\frac{\log(1+\arctan(\frac{x}{2}))}{x}$$ and by l'hospital's $$\log(L)=\lim_{x\rightarrow 0}\frac{1}{1+\arctan(\frac{x}{2})}*\frac{1}{1+(\frac{x}{2})^2}=1\\ \Rightarrow L=e$$
Let $f(x) = 1+\arctan (x/2).$ Apply $\ln $ to the expression of interest to get
$$\frac{\ln (f(x))}{x/2} = 2\frac{\ln f(x) - \ln f(0)}{x-0}.$$
As $x\to 0,$ the last expression $\to 2(\ln f)'(0)$ by definition of the derivative (no L'Hopital used). That's easy enough to compute. Exponentiate back for the original limit.