Use singleID instead of normalID when there is only one sample from a patient

[shutaojccc]

I used to use normalID and tumourID for N-T paired samples, and normalID when the sample is by itself. I later realized that the latter was kind of confusing because tumour sample alone can be used to run call-gSNP as well, and that using singleID when the sample is by itself would make more sense. Not urgent though.

[yashpatel6]

Related:

We use sample_id for the overall sample identifier, ex. sample_id = S00-9422, normal_id = S00-9422N_079_114, tumour_id: S00-9422T_027_166. We should clean up the variable names in this pipeline though, there're some inconsistencies across the processes. I'll keep an issue open for this!

Originally posted by @yashpatel6 in #60 (comment)

Also clean up the inconsistent variable naming.