Changes in version 0.1.0                        

Initial release

  - fapa_core(): SVD-based core profile estimation with sign
    normalisation and variance decomposition.
  - load_and_ipsatize(): CSV loading and row-centering.
  - Three-stage bootstrap verification framework:
      - Stage 1 fapa_pa(): variance-matched Horn's parallel analysis.
      - Stage 2 fapa_procrustes(): subspace stability via Procrustes
        principal angles.
      - Stage 3 fapa_tucker(): profile replicability via Tucker's
        congruence coefficients.
  - fapa_bca(): BCa bootstrap confidence intervals for core-profile
    coordinates via boot::boot.ci().
  - fapa_person(): person-level reconstruction, pattern weights, and
    optional bootstrap CIs for selected participants.
  - Formatted print methods: print_pa(), print_procrustes(),
    print_tucker().
  - Plotting functions: plot_fapa_core(), plot_pa_scree(),
    plot_person_match(), plot_principal_angles(), plot_tucker_cc().
  - CSV output helpers: write_fapa_results(), write_verification().