Add base structure for statistical analysis repository (R, Python, Quarto)

[Copilot]

Bootstraps an empty repo into a functional statistical analysis template for biological/agronomic experimental data, covering frequentist and Bayesian workflows in both R and Python.

Structure added

├── .gitignore                        # R, Python, Quarto, OS/editor artifacts
├── README.md                         # Full docs: model table, requirements, quick-start
├── R/
│   ├── 01_modelos_frecuentistas.R    # ANOVA, LM, LMM, GLM (Poisson/Gamma), GAM, GLMM
│   └── 02_modelos_bayesianos.R       # brms: Bayesian LMM + GLMM-Poisson, LOO comparison
├── Python/
│   ├── 01_modelos_estadisticos.py    # statsmodels + pygam: ANOVA, LM, LMM, GLM, GAM
│   └── 02_modelos_bayesianos.py      # PyMC + ArviZ: Bayesian LM, LMM, GLMM-Poisson
└── quarto_templates/
    ├── reporte_analisis.qmd          # Frequentist report (HTML + PDF)
    └── reporte_bayesiano.qmd         # Bayesian report (HTML + PDF)

Key design decisions

• Parity between R and Python: both languages cover the same model families so teams can work in either ecosystem.
• Bayesian scripts use modern tooling: brms/Stan in R; PyMC + ArviZ in Python — including MCMC diagnostics and LOO-CV model comparison.
• Quarto templates are render-ready: parameterised YAML front-matter, code-fold, embedded resources, and cache/ auto-creation for brms model caching.
• All scripts use simulated data with clear substitution points (read.csv / pd.read_csv) so they run out of the box and are easy to adapt.

Notable fix

rpois lambda recycling bug in R/02_modelos_bayesianos.R: lambda = c(3, 6, 10) was recycling element-wise across all 60 rows instead of assigning one rate per treatment group — corrected to lambda = rep(c(3, 6, 10), each = 20).

Original prompt

Crea la estructura base para un repositorio de análisis estadístico. Incluye una carpeta para plantillas de Quarto (.qmd) orientadas a reportes de análisis de datos. Crea directorios separados para scripts de R y Python. Además, incluye un par de scripts de ejemplo con la estructura básica para correr modelos estadísticos (como ANOVA, LM, LMM, GLM, GAMs, GLMM Y BAYESIANA) aplicados a datos experimentales biológicos o agronómicos. Por último, genera un archivo README.md bien estructurado y un archivo .gitignore estándar para proyectos de R y Python