There are many instances of loco boilers with multiple fire tubes being successfully fired by gas, over in the USA they seem to use it a lot especially in areas prone to forest fires. I gave two references above, here's another USA, and there was an article in SMEE journal some years ago. Also google for Marty burners, ME 3169 and Engineering in Miniature May 2011. ME 4188 had an article on using a modification of a Sievert cyclone burner in a 5"g metro and a 3.5" Rob Roy. Even LBSC showed drawings of a primus stove burner in a model loco firebox, no idea if it worked but his stuff tended to.
Where I will agree wholeheartedly with Fizzy is that you are unlikely to have success in a muti-tube boiler without forced draught, too much pressure drop in the tubes. What you also need to ensure is that the gas has completely burned before it gets into the tubes, once inside it is cooled faster than the burning can heat it, the flame is extinguished and you get unburnt or partially burnt gas coming out of the chimney. (This is a bit of a simplification, but I believe basically correct)
I have a suspicion (no more) that you won't get the same output from gas as you will with coal. My vertical boiler has a 4" grate, so allowing 25 lb/sqft/hr grate loading (actual burning) gives 2 lb/hr, 28362 BUT/hr or 8 kW, the same as a Sievert 2941 burner, but you won't get that to fully burn in a firebox only 4" or so high
Finally there is a suggestion, and I can't put it higher than that, that a vapourised paraffin flame is more compact than an equivalent LPG flame. I have no hard data on that, just something someone told me. All we need now is for someone to take all this info and make a successful burner for a vertical boiler instead of my pontifications