[NCCU-Schultz-Lab]

Two pieces of chemistry bookkeeping were written out in several places and had already drifted once. The SLURM resource estimate counted Mn, Co, Ni and Mo as 0 electrons.

Changes

• config.electron_count(atoms, charge) is now the only electron count. Molecule.get_electron_count, molecule.suggest_multiplicity, xyz_input.electron_count and the SLURM estimate all use it. Symbols are title-cased, so suggest_multiplicity now also counts lowercase symbols. The standard-library quantui-batch launcher keeps its own copy, which the existing parity tests hold equal.
• The Calculate tab's solvent-capable calc types are derived from the batch worker's SOLVENT_CALC_TYPES, plus Transition State. TS runs only in the app until it is added to SLURM, so it is the one app-only entry. This replaces a second hand-written list; the result is the same six calc types.

Testing

• pytest -m "not network": 3817 passed, 19 skipped, 0 failed.
• ruff, black and mypy (run as in CI): clean.
• New tests:
  • electron counts for neutral, charged, transition-metal and lowercase input;
  • every caller agrees on MnO₄⁻ (58 electrons);
  • the app's solvent set equals the batch set plus Transition State.

Authorship

• Claude (Opus 5.5): code edits, review, and conceptual discussion
• Jonathan Schultz: overall vision, planning, review, and orchestration