Try to downsample data in Fiji BDV Playground to match Atlas resolution

[lacan]

That wasy brainreg would not need to resample stuff, which should make it faster, especially as brainreg parameters should sometimes be optimized and this would reduce needless overhead lined to #5

[lacan]

Also need to add the atlas resolution to the parameters of brainreg in the YML

[abisi]

Hi! I think we're okay with brainreg downsampling data at the atlas resolution ?
Also, the atlas resolution is part of the name of brainglobe atlas. So maybe we can close this as it is taken care of?