At 1.5A, I don't see how a wire resistance at 2 meters would be a problem. It's all about resistive losses. Say, overkilling it with 16AWG (or 1.5mm^2), you are already in near-zero loss territory. The motor windings are very thin and long in comparison. The controller IC specified at 3A continuous (assumably with good heatsinking practices, i.e. proper thermal vias to large copper planes, and good ventilation around the PCB), I see you have a good margin if you only need 1.5A.
Seeing the IC is rated for 500kHz(!), this means that at any sensible switching frequency (say 25kHz), the switching losses must be a tiny portion of total losses, so I^2R losses dominate. As you are using half of the max rated current, the losses are one quarter of what the IC is designed for.
The controller has actual real cycle-by-cycle current limit as it definitely should, so you have picked a good part in this sense. In theory, this seems a good choice. I have no experience with this particular part so can't say for sure, TI sometimes have serious bugs in their parts rendering them totally unusable, so I'm always a bit scared when choosing any complex IC. Evaluate it first properly before committing in the design.
Having a tight loop is paramount, but the critical loop is between the input capacitor and the two switches (two mosfets, or one mosfet plus diode). This loop must be minimized, and the integrated power switches make that very easy, the only thing you need to think about is DC bus capacitance. Having several power/GND pins on the device helps, you can put a ceramic capacitor next to each pair, and a bigger electrolytic further away for damping parasitic LC resonances. Output wires only add a bit of inductance in series with the motor inductance, which is already huge (and only beneficial, for current regulation that is).
Really excessive lead inductance can increase the voltage spikes the motor sees, but I don't think this is a problem here.
All this being said, output wires can still radiate EMI; not only from the driver switching, but from the brush arcing as well; in this sense, the closer you get the driver, the better.
If you add filters or snubbers to deal with brush noise, note they need to be able to withstand the PWM control while not drawing excessive AC currents from the switching.