We shall use the Feynman trick. Instead of the integral in question
$$I=\int_{0}^{\infty}\frac{\log(x)}{e^x+1}\;dx$$
consider the integral
$$i(s)=\int_{0}^{\infty}\frac{x^{s-1}}{e^x+1}\;dx$$
We can generate the log by taking the drivative with respect to $s$ at $s=1$, indeed
$$I = \frac{d}{ds}i(s)|_{s\to 1}$$
Writing
$$\frac{x^{s-1}}{e^x+1}=\frac{e^{-x} x^{s-1}}{1+e^{-x}}=- x^{s-1}\sum_{n=1}^ {\infty} (-1)^n e^{-n x}$$
the integral of the summand becomes
$$\int_{0}^{\infty}x^{s-1}e^{-n x}\;dx \overset{nx\to y}=\frac{1}{n^s}\int_{0}^{\infty}y^{s-1}e^{-y}\;dy=\frac{\Gamma(s)}{n^s}$$
summing over $n$ gives the integral
$$i(s) = \Gamma(s)\sum_{n=1}^ {\infty} (-1)^{n+1} \frac{1}{n^s}=\Gamma(s)\left(1-2^{1-s}\right) \zeta (s)$$
Now we calculate the derivative, close to $s=1$
$$\begin {align}&\frac{d}{ds}\left(\Gamma(s)\left(1-2^{1-s}\right) \zeta (s)\right) \\& =\left(\Gamma'(s)\left(1-2^{1-s}\right) \zeta (s)\right)+\left(\Gamma(s)\left(1-2^{1-s}\right)' \zeta (s)\right)+\left(\Gamma(s)\left(1-2^{1-s}\right) \zeta' (s)\right)\\
&\simeq \left(-\gamma \log (2)\right)+\left(\frac{\log (2)}{s-1}-\log ^2(2)\right)+\left(\frac{\log ^2(2)}{2}+\gamma \log (2)-\frac{\log (2)}{s-1}\right)+O\left((s-1)^1\right)\\
&= -\log ^2(2)+\left(\frac{\log ^2(2)}{2}\right)+O\left((s-1)^1\right)\\
&\overset{s \to 1}
=
-\frac{1}{2}\log ^2(2)\end{align}$$
which is the desired result.
Discussion
Consider more generally
$$I_n = \int_0^{\infty } \frac{\log ^n(x)}{e^x+1} \, dx$$
It is given by the $n$-th derivative of $i(s)$ at $s=1$.
We have for instance
$$I_2 = \frac{\log ^3(2)}{3}-2 \gamma _1 \log (2)+\gamma ^2 \log (2)+\frac{1}{6} \pi ^2 \log (2)$$
In general in $I_n$ there appears a term $\frac{\log ^{n+1}(2)}{n+1}$. Can you prove this?