Snubber cap and R

I agree with the TVS as this quickly discharges the energy with high V, low current. But watch out for parasitic inductance in the loop with TVS. I recommend the Snubberless Alternistors. 600V or 800V https://www.st.com/en/thyristors-scr-and-ac-switches/standard-and-snubberless-triacs.html#products

If you are doing this EMI reasons, then a slower response is needed with a resistor and maybe a cap

 

I love Falstad's ability to use your imagination. If you can scheme , you can simulate it.

Actually, your RC snubber looks best for EMI and OVP except you draw twice the power. No big deal. TVS is good for Triac protection and faster actuator release
(Tau) but that also generates EMI.

Below I simulated a Triac with ZCS turn off by controlling phase and I controlled turn on time with duty cycle . THen I put in ideal SPDT switch to compare 3 methods in one scope trace ( slow-mo)

Triac noise is on the bottom trace and the control voltage was on a logic control pulse with 90% duty cycle to a switch with 10 mohms (not 100m) So I clicked on each switch during a slow sweep to capture each trace to compare the results. I assume I have shown optimal config. But you can play with this too if you want the simulation link.
Bl;

To be clear, the TVS is better for conducted noise to grid. but worse for local radiation near Triac as it switches fast to the clamp voltage. But you can't put a cap like I did across pulsed Switch(Triac sim) as the coil would conduct when off.

 

If you examine my bottom two traces, the noise around the triac and the noise generated on the grid can be easily simulated with a loop of wire and a small relay coil with 50% V to the coil in series with the NC contacts. THis will radiate enough noise to see if your breadboard or protoboard will be immune to interference. The RC snubber is better around the Triac and the TVS with cap is almost as good and better on the grid but it must be a Y cap.