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>

DESCRIPTION

@@ -0,0 +1,34 @@
+Package: baflakehouse
+Title: Lakehouse Workflow for the Bank Account Fraud Dataset
+Version: 0.0.0.9000
+Authors@R: 
+    person("Rob", "Wiederstein", role = c("aut", "cre"),
+           email = "REPLACE_ME@example.com")
+Description: Tools to ingest the Bank Account Fraud (BAF) Base dataset into a
+    MinIO/S3-backed lakehouse, clean encoded missing values, and produce
+    reproducible reporting artifacts (tables, figures, slides) orchestrated with
+    targets.
+License: MIT + file LICENSE
+Encoding: UTF-8
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.3.3
+Imports:
+    arrow,
+    colorspace,
+    cowplot,
+    dplyr,
+    tidyr,
+    stringr,
+    readr,
+    gt,
+    quarto,
+    ggplot2,
+    bonsai
+Suggests:
+    duckdb,
+    targets,
+    tarchetypes,
+    knitr,
+    scales
+URL: https://docs.robwiederstein.org/baflakehouse
+BugReports: https://git.robwiederstein.org/rkw/bank-fraud-baf-lakehouse/issues