Recently, advances in film synthesis methods have enabled a study of extremely overdoped $La_{2-x}Sr_{x}CuO_{4}$. This has revealed a surprising behavior of the superfluid density as a function of doping and temperature, the explanation of which is vividly debated. One popular class of models posits electronic phase separation, where the superconducting phase fraction decreases with doping, while some competing phase (e.g. ferromagnetic) progressively takes over. A problem with this scenario is that all the way up to the dome edge the superconducting transition remains sharp, according to mutual inductance measurements. However, the physically relevant scale is the Pearl penetration depth, $Lambda_{P}$, and this technique probes the sample on a length scale $L$ that is much larger than $Lambda_{P}$. In the present paper, we use local scanning SQUID measurements that probe the susceptibility of the sample on the scale $L << Lambda_{P}$. Our SQUID maps show uniform landscapes of susceptibility and excellent overall agreement of the local penetration depth data with the bulk measurements. These results contribute an important piece to the puzzle of how high-temperature superconductivity vanishes on the overdoped side of the cuprates phase diagram.