spec: add roc-curve specification (#2278)

## New Specification: `roc-curve`

Related to #2273

---

### specification.md

# roc-curve: ROC Curve with AUC

## Description

A Receiver Operating Characteristic (ROC) curve visualizes the
performance of a binary classifier by plotting the True Positive Rate
(TPR) against the False Positive Rate (FPR) at various classification
thresholds. The Area Under the Curve (AUC) provides a single metric
summarizing model performance, where 1.0 indicates perfect
classification and 0.5 represents random guessing.

## Applications

- Evaluating binary classification model performance in machine learning
pipelines
- Comparing multiple models to select the best classifier for production
- Selecting optimal classification thresholds based on
sensitivity/specificity trade-offs
- Assessing diagnostic test accuracy in medical research

## Data

- `fpr` (numeric) - False Positive Rate values (0 to 1)
- `tpr` (numeric) - True Positive Rate values (0 to 1)
- `auc` (numeric) - Area Under the Curve score (0 to 1)
- Size: typically 100-1000 threshold points per curve
- Example: sklearn.metrics.roc_curve output from binary classification
predictions

## Notes

- Include a diagonal reference line (y=x) representing random classifier
performance
- Display AUC score in legend or annotation
- Use distinct colors/styles when comparing multiple models
- Axes should range from 0 to 1 with equal aspect ratio preferred

---

**Next:** Add `approved` label to the issue to merge this PR.

---
:robot: *[spec-create
workflow](https://github.com/MarkusNeusinger/pyplots/actions/runs/20526502674)*

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

Author: github-actions[bot]

plots/roc-curve/specification.md

@@ -0,0 +1,27 @@
+# roc-curve: ROC Curve with AUC
+
+## Description
+
+A Receiver Operating Characteristic (ROC) curve visualizes the performance of a binary classifier by plotting the True Positive Rate (TPR) against the False Positive Rate (FPR) at various classification thresholds. The Area Under the Curve (AUC) provides a single metric summarizing model performance, where 1.0 indicates perfect classification and 0.5 represents random guessing.
+
+## Applications
+
+- Evaluating binary classification model performance in machine learning pipelines
+- Comparing multiple models to select the best classifier for production
+- Selecting optimal classification thresholds based on sensitivity/specificity trade-offs
+- Assessing diagnostic test accuracy in medical research
+
+## Data
+
+- `fpr` (numeric) - False Positive Rate values (0 to 1)
+- `tpr` (numeric) - True Positive Rate values (0 to 1)
+- `auc` (numeric) - Area Under the Curve score (0 to 1)
+- Size: typically 100-1000 threshold points per curve
+- Example: sklearn.metrics.roc_curve output from binary classification predictions
+
+## Notes
+
+- Include a diagonal reference line (y=x) representing random classifier performance
+- Display AUC score in legend or annotation
+- Use distinct colors/styles when comparing multiple models
+- Axes should range from 0 to 1 with equal aspect ratio preferred

plots/roc-curve/specification.yaml

@@ -0,0 +1,31 @@
+# Specification-level metadata for roc-curve
+# Auto-synced to PostgreSQL on push to main
+
+spec_id: roc-curve
+title: ROC Curve with AUC
+
+# Specification tracking
+created: 2025-12-26T17:27:56Z
+updated: 2025-12-26T17:27:56Z
+issue: 2273
+suggested: MarkusNeusinger
+
+# Classification tags (applies to all library implementations)
+# See docs/concepts/tagging-system.md for detailed guidelines
+tags:
+  plot_type:
+    - line
+    - roc
+    - curve
+  data_type:
+    - numeric
+    - continuous
+  domain:
+    - statistics
+    - machine-learning
+    - healthcare
+  features:
+    - basic
+    - evaluation
+    - classification
+    - comparison