Remove factorial assembly workflow from the biological interaction graph #1727

CommunityMech:000399; PMID:42153623. Remove only the full-factorial assembly-method node. Preserve the existing measured phenotype and statistical interpretation, qualifying the illustrative ranking and nonlinear absorbance-to-biomass relation. The authors explicitly distinguish statistical interaction from a resolved biological mechanism. Do not add biological arrows or actionable pathogen growth/fitness optimization. Five panel members will lose artificial connectivity supplied by the method node; leave those warnings visible pending separate panel-identity review. Fresh abstract and existing open-access cached primary results/discussion read; fresh full text failed. Close only after protected-main merge.

All-state issue titles/bodies searched by IDs, references and stems. Ignored/hidden local search included, excluding .git/docs/site/references_cache; relevant caches separately inspected. Closed #1189-#1192 and earlier validation/rendering issues do not cover these findings. Existing #765 owns broader propanotroph re-evidencing after #754; no duplicate non-graph issue filed for that scope. Comments and PRs not exhaustively searched. Source-based self-adversarial review, not independent approval.