I ohm'd the burned board and it looks like there's a short from Vin to SW1/2/3. I didn't test this resistance before running the board so I don't know if it failed to a short or the short pre-existed.
(bolded lines are nodes likely to short due to pin proximity under IC)
I feel like it's unlikely this was a short due to solder under the IC because it would require a very unlikely solder blob to form and NOT short anything to GND.
I'm going to assemble another and do this kind of meticulous resistance check before applying power.
Thanks Benno. You don't think the electrolytic would have sufficient ESR on its own, you needed to add series resistance to the ceramics?
On the specific PCB there was no room for an electrolytic cap.
So the only solution was to put a series resistor en the ceramic to prevent ringing.
Later revisions I added a electrolytic and kept the series resistor.
Benno
Thanks everyone for their help, I finally got this working. It was assembly issues I think, since I kept desoldering and resoldering the IC several times getting different results each time (even though no solder shorts ever showed on my DMM). On the zillionth try I got working 5V output.
I have some work to do on soldering technique.
FWIW the design starts perfectly fine from a current limited low input voltage, slowly ramping it up, with no load. The input current is very low (only ~3mA) when done this way. Now I know what to look for I can bring up additional prototypes safely.