Initial commit: BAF Lakehouse fraud detection pipeline

End-to-end LightGBM fraud detection pipeline built as an R package,
orchestrated by targets with data stored in MinIO via Apache Arrow.
Includes 6-layer Lakehouse architecture, class imbalance tournament,
formally tuned hyperparameters (PR-AUC 0.198), and Quarto RevealJS slides.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

man/plot_fraud_by_month.Rd

@@ -0,0 +1,34 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{plot_fraud_by_month}
+\alias{plot_fraud_by_month}
+\title{Plot applications by month (Legit vs Fraud) on a log scale}
+\usage{
+plot_fraud_by_month(
+  dataset_prefix,
+  bucket_name = "baf-fraud",
+  palette = "Dark 3",
+  title = ""
+)
+}
+\arguments{
+\item{dataset_prefix}{Character. Prefix inside the bucket, e.g. "03_primary/variant=Base".}
+
+\item{bucket_name}{Character. Bucket name. Default "baf-fraud".}
+
+\item{palette}{Character. colorspace qualitative palette name. Default "Dark 3".}
+
+\item{title}{Character. Plot title. Default "".}
+}
+\value{
+A ggplot object.
+}
+\description{
+Builds an exploratory chart of absolute application counts by month
+split by outcome (Legit vs Fraud). Uses a log10 y-axis so rare fraud
+remains visible on the same axis.
+}
+\details{
+Data source: expects a cleaned "primary" dataset prefix (e.g. 03_primary/variant=Base)
+stored in MinIO/S3, accessed via \code{connect_baf()}.
+}