If a likelihood function $L(\theta)$ is given by $L(\theta) \propto \exp(-\frac{1}{2}(\theta - 3)^2)$, and we assume a flat prior $\pi(\theta) \propto 1$, what is the posterior distribution $\pi(\theta|x)$?