I'm posting my whole thought process, but I'd like to ask specifically about whether is my expansion of $\log(1+y)$ into Taylor series good & allowed in this situation? Is there an easier (not saying this isn't an easy way) to compute this limit?
$$\\ \lim_{x\rightarrow \infty}x(e-(1+\frac{1}{x})^x) = \lim_{y \rightarrow 0}\frac{e-(1+y)^{\frac{1}{y}}}{y} = \lim_{y \rightarrow 0}\ \frac{e-e^\frac{\log(1+y)}{y}}{y} = \lim_{y \rightarrow 0} \ \frac{e-e^\frac{y-\frac{y^2}{2}+O(y^3)}{y}}{y} = \\ =\lim_{y \rightarrow 0}\ \frac{e-e^{1-\frac{y}{2}+O(y^2)}}{y}= e\lim_{y \rightarrow 0}\ \frac{1-e^{-\frac{y}{2}+O(y^2)}}{y} = (-e)\lim_{y \rightarrow 0}\ \frac{e^{-\frac{y}{2}+O(y^2)} - 1}{y} =\\ (-e)\lim_{y \rightarrow 0}\ \frac{e^{y (\frac{-1}{2})+O(y^2)} - 1}{y}$$
And now, from $$ \lim_{y \rightarrow 0}\ \frac{e^{ay} - 1}{y} = a$$
Follows
$$ (-e)*(\frac{-1}{2}) = \frac{e}{2}$$