bulkMAE 0.4.1

• Records Fan XingFu as the package author and maintainer.
• Stops pkgdown reference builds from following GSVA S4 Rd links, and keeps
  the pkgdown library uncached so a broken magick stack cannot abort the
  site build.
• Adds bulkmae_rename() and a vignette/pkgdown copy of the 0.4 naming map
  so old run_* / infer_* / prepare_* calls fail with an explicit
  replacement.
• Runs BiocCheck on main.
• Documents installation from the GitHub Release tag or attached source
  tarball, and keeps portability, full-backend, and coverage off pull
  requests.

bulkMAE 0.4.0

• Replaces mixed run_*, infer_*, prepare_*, and backend-specific naming
  with discoverable <family>_<method>() / <family>_<operation>() APIs.
• Organizes public functions under mae_*, qc_*, reduce_*, de_*,
  enrich_*, score_*, activity_*, cluster_*, coexpr_*, network_*,
  deconv_*, surv_*, ml_*, meta_*, and drug_* families.
• Consolidates GO, KEGG, and Reactome ORA/GSEA combinations behind explicit
  method arguments instead of multiplying function names.
• Removes old exported aliases intentionally so autocomplete exposes one
  coherent API; an installed migration map records every breaking rename.
• Restores compatibility across the declared R 4.4+ dependency range, including
  edgeR tximport/differential-splicing fallbacks, current dream moderation,
  timeROC namespace resolution, decoupleR method-argument conventions, and
  namespace-only NMF, WGCNA, MuSiC, and immunedeconv execution.
• Adds explicit MAE assay/feature helpers and direct TPM normalization from gene
  lengths stored in rowData, an aligned assay, or a supplied vector/matrix.
• Completes a simulated-data suite spanning all 84 public functions, including
  real offline xCell, MuSiC, BayesPrism, NMF, WGCNA, GO, and Reactome backend
  smoke tests without requiring network services.
• Keeps network-backed tests disabled by default. Set
  BULKMAE_RUN_ONLINE_TESTS=true only when deliberately testing remote resources.
• Extends the standard ggplot2 layer with statistically faithful GSEA
  classic/ridge plots and ORA bubble/community/radial plots. These views keep
  rank vectors, ORA ratios, enriched-feature membership, community Jaccard
  overlap, and radial shared-feature counts distinct instead of coercing them
  into one generic effect. ORA views reproduce the reference term-specific
  community nodes, unique radial nodes, compact adj p legend, and white
  publication backgrounds.
• Adds a concise executable Chinese airway enrichment tutorial covering
  method selection, native GO ORA/GSEA results, and all five enrichment plots.