feat(altair): implement contour-density (#2585)

## Implementation: `contour-density` - altair

Implements the **altair** version of `contour-density`.

**File:** `plots/contour-density/implementations/altair.py`

---
:robot: *[impl-generate
workflow](https://github.com/MarkusNeusinger/pyplots/actions/runs/20593353049)*

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

Author: github-actions[bot]

plots/contour-density/implementations/altair.py

@@ -0,0 +1,120 @@
+"""pyplots.ai
+contour-density: Density Contour Plot
+Library: altair 6.0.0 | Python 3.13.11
+Quality: 88/100 | Created: 2025-12-30
+"""
+
+import altair as alt
+import numpy as np
+import pandas as pd
+from scipy.stats import gaussian_kde
+
+
+# Data - simulating temperature and humidity measurements showing natural clusters
+np.random.seed(42)
+
+# Create three distinct clusters representing different climate conditions
+n1 = 150
+temp1 = np.random.normal(15, 4, n1)
+humidity1 = np.random.normal(30, 8, n1)
+
+n2 = 200
+temp2 = np.random.normal(25, 5, n2)
+humidity2 = np.random.normal(55, 10, n2)
+
+n3 = 150
+temp3 = np.random.normal(38, 4, n3)
+humidity3 = np.random.normal(75, 8, n3)
+
+# Combine data
+temperature = np.concatenate([temp1, temp2, temp3])
+humidity = np.concatenate([humidity1, humidity2, humidity3])
+
+# Compute 2D KDE for density estimation
+xy = np.vstack([temperature, humidity])
+kde = gaussian_kde(xy)
+
+# Create grid for density estimation
+n_grid = 60
+x_grid = np.linspace(temperature.min() - 5, temperature.max() + 5, n_grid)
+y_grid = np.linspace(humidity.min() - 5, humidity.max() + 5, n_grid)
+xx, yy = np.meshgrid(x_grid, y_grid)
+positions = np.vstack([xx.ravel(), yy.ravel()])
+z = kde(positions).reshape(xx.shape)
+
+# Create filled contour visualization using binned density levels
+n_levels = 10
+z_min, z_max = z.min(), z.max()
+levels = np.linspace(z_min, z_max, n_levels + 1)
+z_binned = np.digitize(z, levels) - 1
+z_binned = np.clip(z_binned, 0, n_levels - 1)
+
+# Create grid cell data for filled contours
+step_x = x_grid[1] - x_grid[0]
+step_y = y_grid[1] - y_grid[0]
+
+grid_data = pd.DataFrame({"x": xx.ravel(), "y": yy.ravel(), "density_level": z_binned.ravel()})
+
+# Create filled contour chart using mark_rect with binning
+x_domain = [float(temperature.min() - 6), float(temperature.max() + 6)]
+y_domain = [float(humidity.min() - 6), float(humidity.max() + 6)]
+
+filled_contours = (
+    alt.Chart(grid_data)
+    .mark_rect()
+    .encode(
+        x=alt.X(
+            "x:Q",
+            bin=alt.Bin(step=step_x),
+            scale=alt.Scale(domain=x_domain),
+            title="Temperature (°C)",
+            axis=alt.Axis(labelFontSize=18, titleFontSize=22, grid=True, gridOpacity=0.3),
+        ),
+        y=alt.Y(
+            "y:Q",
+            bin=alt.Bin(step=step_y),
+            scale=alt.Scale(domain=y_domain),
+            title="Humidity (%)",
+            axis=alt.Axis(labelFontSize=18, titleFontSize=22, grid=True, gridOpacity=0.3),
+        ),
+        color=alt.Color(
+            "mean(density_level):Q",
+            scale=alt.Scale(scheme="blues"),
+            title="Density",
+            legend=alt.Legend(titleFontSize=20, labelFontSize=16, gradientLength=400, gradientThickness=25),
+        ),
+    )
+)
+
+# Add contour line effect by layering darker outlines at level boundaries
+contour_data = []
+for i in range(n_grid - 1):
+    for j in range(n_grid - 1):
+        current = z_binned[i, j]
+        # Check if adjacent cells have different levels (boundary)
+        if i < n_grid - 1 and z_binned[i + 1, j] != current:
+            contour_data.append({"x": x_grid[j], "y": (y_grid[i] + y_grid[i + 1]) / 2, "level": current})
+        if j < n_grid - 1 and z_binned[i, j + 1] != current:
+            contour_data.append({"x": (x_grid[j] + x_grid[j + 1]) / 2, "y": y_grid[i], "level": current})
+
+if contour_data:
+    contour_df = pd.DataFrame(contour_data)
+    contour_points = (
+        alt.Chart(contour_df)
+        .mark_circle(size=8, opacity=0.6, color="#1a4d80")
+        .encode(x=alt.X("x:Q", scale=alt.Scale(domain=x_domain)), y=alt.Y("y:Q", scale=alt.Scale(domain=y_domain)))
+    )
+    chart = alt.layer(filled_contours, contour_points)
+else:
+    chart = filled_contours
+
+# Configure the chart with proper sizing
+chart = chart.properties(
+    width=1600, height=900, title=alt.Title(text="contour-density · altair · pyplots.ai", fontSize=28, anchor="middle")
+).configure_view(strokeWidth=0)
+
+# Save as PNG (1600 * 3 = 4800, 900 * 3 = 2700)
+chart.save("plot.png", scale_factor=3.0)
+
+# Save as interactive HTML
+chart.save("plot.html")

plots/contour-density/metadata/altair.yaml

@@ -0,0 +1,25 @@
+library: altair
+specification_id: contour-density
+created: '2025-12-30T09:33:23Z'
+updated: '2025-12-30T10:27:05Z'
+generated_by: claude-opus-4-5-20251101
+workflow_run: 20593353049
+issue: 0
+python_version: 3.13.11
+library_version: 6.0.0
+preview_url: https://storage.googleapis.com/pyplots-images/plots/contour-density/altair/plot.png
+preview_thumb: https://storage.googleapis.com/pyplots-images/plots/contour-density/altair/plot_thumb.png
+preview_html: https://storage.googleapis.com/pyplots-images/plots/contour-density/altair/plot.html
+quality_score: 88
+review:
+  strengths:
+  - Good visual design with blues color scheme providing clear density gradient
+  - Proper axis labels with units (Temperature °C, Humidity %)
+  - Realistic climate data with three distinct clusters demonstrating density patterns
+  - Well-proportioned layout with balanced margins and proper legend placement
+  - Correct title format following pyplots.ai convention
+  weaknesses:
+  - Contour lines (mark_rule) are not visible in the output - the filled density background
+    renders but the isoline overlay does not appear
+  - Code violates KISS principle by using custom extract_contours function instead
+    of simpler approach