I do not know if it is true when the residue field is finite, but let me show you why it holds if we assume that $R/\mathfrak{m}$ is infinite.
First note that the condition $\mathfrak{m}^2=(\overline{x})\mathfrak{m}$ for some regular sequence $\overline{x}$ is equivalent to say that $\mathfrak{m}^2=I\mathfrak{m}$ for some minimal reduction $I$ of $\mathfrak{m}$, because a minimal reduction is generated by $\dim R$ elements that are a regular sequence. Consider a minimal reduction $I$ of $\mathfrak{m}$. The length of the $R$-module $R/\mathfrak{m}I$ can be computed in these two ways:
\begin{align}
\ell_R(R/\mathfrak{m}I)&=\ell_R(R/\mathfrak{m})+\ell_R(\mathfrak{m}/\mathfrak{m}^2)+ \ell_R(\mathfrak{m^2}/\mathfrak{m}I)=1+\mu(\mathfrak{m})+\ell_R(\mathfrak{m^2}/\mathfrak{m}I), \\
\ell_R(R/\mathfrak{m}I)&= \ell_R(R/I)+ \ell_R(I/\mathfrak{m}I)=e(R)+ \dim(R).
\end{align}
Hence, it follows that $\mu(\mathfrak{m})-\dim(R)+1+\ell_R(\mathfrak{m}^2/\mathfrak{m}I)=e(R)$ and $R$ has minimal multiplicity if and only if $\mathfrak{m}^2=\mathfrak{m} I$.
If the residue field is finite and $\mathfrak{m}^2=(\overline{x})\mathfrak{m}$ for some regular sequence $\overline{x}$, the proof above shows that $R$ has minimal multiplicity, but I do not believe that the converse is true.