As much as possible of the code uses MSnbase objects to keep data. For
example, instead of CSV files, the chromatograms are now inside a
Chromatograms object.

Things have been adapted to this until the preproc. The preproc stage
is still not working properly for MS2 part.

Logic in separating unknowns and knows was faulty.

R/mix.R (vald_comp_tab): New function. It checks for the existence of
ID, Name, RT(rt), SMILES and mz as needed in compound lists. It also
checks if IDs are unique, and if when SMILES, or mz are given, they
are not NAs.

Clarified the rtDelta description in QA box to mean RT shift tolerance
and added rtDeltaWin widget to spectra extraction box.

Retention time functionality in chromatograms.

Some naming changes.

R/extraction.R (gen_rt_range): Generates rt range matrix given rt-s
and tolerance.

This is where constants go.

The problem seems to have been with the ordering in the
verif_prec_fine procedure.

More to follow.

But, MS2 are way off

Also, move `...prescreen_df' to extraction.R and rename it to
extr_rmb. Create driver level `extract' function that will be used to
call the backend extraction function based on user's choice.