Including an explanation in tutorials of DADA2/feature-classifier long run times & job splitting

[mestaki]

Improvement Description
A common enough forum topic is regarding DADA2/Deblur/feature-classifier run times. Users follow the tutorials and expect the same short run times on their own data. A brief explanation of factors affecting run-times and especially highlighting the --n-thread parameter in the tutorial might be useful and not get people's expectations too high:P Perhaps even include --n-threads 1 in the tutorial and then explain to change that to whatever depending on your machine. Having it in there will make users look into it perhaps.

References
ex forum topic