tacular-omics/tacular: v1.0.1 - Zenodo + Docs

Add tacular update CLI to refresh ontology data from the latest OBO releases at runtime, cached per-user and preferred over bundled data. Fix isotope-labelled atoms (13C, 15N, ...) being dropped from composition/formula while the mass stayed correct. Fix repeated element symbols collapsing in neutral-loss/fragment-ion formula parsing (e.g. formic acid, formamide). Refresh bundled UNIMOD data to the latest release. Fix all 9 internal fragment ion offsets, which were systematically wrong (the table was shifted so by was -CO instead of 0). Corrected to internal(F,B) = δF + δB - H2O, consistent with the a/b/c/x/y/z offsets. Single-source the OBO parsing logic (shared by the build pipeline and tacular update). Improve logging/error diagnosability: _datagen parser failures now log the exception type/message and a full traceback (not just "something failed"), and a corrupt runtime cache logs via logging instead of warnings.warn so it can never raise under -W error. Add -v/-vv verbosity flags to the tacular CLI. Add CLAUDE.md/AGENTS.md/llms.txt for AI coding agents and LLM tooling; fill in missing module/function docstrings across the public API; correct stale claims in data_gen/README.md. Add a UNIPROT_PTM_LOOKUP lookup for UniProt's controlled vocabulary of posttranslational modifications (ptmlist.txt), with tacular update/cache support like the OBO ontologies. Cross-references to PSI-MOD/UNIMOD are resolvable via get_psimod()/get_unimod().