Omit `tolerance_ppm` and each experiment gets the window its own nucleus
calls for: 1H 0.3 ppm, 13C 3.0.

The carbon-edited slots are separate because the experiments say different
things. An APT shows every carbon and separates them by the PARITY of the
proton count, so it cannot tell CH from CH3 or Cq from CH2. A DEPT-135 shows
only protonated carbons: a line in it proves the carbon carries protons, a
line on the CH2 side is exactly CH2, and a carbon that appears in `path_13c`
but not here has exactly zero — a claim no APT can make. A DEPT-90 shows CH
only and splits CH from CH3.

Supply a DEPT-135 together with a plain 13C whenever you have both: the
quaternary carbons come from the difference between them, and a DEPT alone
cannot supply the carbon peak list.

The file decides which sequence it is, from seqfil/PULPROG — a spectrum in
the wrong slot is reported in experiment_warnings and analysed as what it
actually is, never as what the slot claimed.