feat(altair): implement histogram-density (#2476)

## Implementation: `histogram-density` - altair

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

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

---
:robot: *[impl-generate
workflow](https://github.com/MarkusNeusinger/pyplots/actions/runs/20584328879)*

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Author: github-actions[bot]

plots/histogram-density/implementations/altair.py

@@ -0,0 +1,76 @@
+""" pyplots.ai
+histogram-density: Density Histogram
+Library: altair 6.0.0 | Python 3.13.11
+Quality: 93/100 | Created: 2025-12-29
+"""
+
+import altair as alt
+import numpy as np
+import pandas as pd
+from scipy import stats
+
+
+# Data - Generate bimodal distribution to show density histogram features
+np.random.seed(42)
+# Reaction times from two conditions: baseline and fatigued
+baseline_times = np.random.normal(loc=250, scale=30, size=350)
+fatigued_times = np.random.normal(loc=380, scale=45, size=150)
+reaction_times = np.concatenate([baseline_times, fatigued_times])
+
+# Compute histogram data manually for density normalization
+num_bins = 25
+bins = np.linspace(reaction_times.min() - 10, reaction_times.max() + 10, num_bins + 1)
+counts, bin_edges = np.histogram(reaction_times, bins=bins, density=True)
+bin_width = bin_edges[1] - bin_edges[0]
+
+# Create DataFrame with bin ranges for proper bar rendering
+hist_df = pd.DataFrame(
+    {
+        "bin_start": bin_edges[:-1],
+        "bin_end": bin_edges[1:],
+        "Density": counts,
+        "bin_center": (bin_edges[:-1] + bin_edges[1:]) / 2,
+    }
+)
+
+# Create density histogram using rect mark for proper filled bars
+histogram = (
+    alt.Chart(hist_df)
+    .mark_rect(color="#306998", opacity=0.75, stroke="#1a3a5c", strokeWidth=1.5)
+    .encode(
+        x=alt.X("bin_start:Q", scale=alt.Scale(domain=[bins.min(), bins.max()]), title="Reaction Time (ms)"),
+        x2="bin_end:Q",
+        y=alt.Y("Density:Q", scale=alt.Scale(domain=[0, counts.max() * 1.1]), title="Density (probability per ms)"),
+        tooltip=[
+            alt.Tooltip("bin_center:Q", title="Bin Center", format=".0f"),
+            alt.Tooltip("Density:Q", title="Density", format=".5f"),
+        ],
+    )
+)
+
+# Create KDE overlay for theoretical density reference
+kde = stats.gaussian_kde(reaction_times, bw_method=0.15)
+x_kde = np.linspace(reaction_times.min() - 20, reaction_times.max() + 20, 300)
+y_kde = kde(x_kde)
+
+kde_df = pd.DataFrame({"Reaction Time (ms)": x_kde, "Density": y_kde})
+
+kde_line = alt.Chart(kde_df).mark_line(color="#FFD43B", strokeWidth=4).encode(x="Reaction Time (ms):Q", y="Density:Q")
+
+# Combine histogram and KDE
+chart = (
+    alt.layer(histogram, kde_line)
+    .properties(
+        width=1600,
+        height=900,
+        title=alt.Title("histogram-density · altair · pyplots.ai", fontSize=28, anchor="middle", color="#333333"),
+    )
+    .configure_axis(labelFontSize=18, titleFontSize=22, gridColor="#cccccc", gridOpacity=0.3)
+    .configure_view(strokeWidth=0)
+)
+
+# Save as PNG (1600 × 900 × 3 = 4800 × 2700 px)
+chart.save("plot.png", scale_factor=3.0)
+
+# Save interactive HTML version
+chart.save("plot.html")

plots/histogram-density/metadata/altair.yaml

@@ -0,0 +1,25 @@
+library: altair
+specification_id: histogram-density
+created: '2025-12-29T22:49:29Z'
+updated: '2025-12-29T22:58:35Z'
+generated_by: claude-opus-4-5-20251101
+workflow_run: 20584328879
+issue: 0
+python_version: 3.13.11
+library_version: 6.0.0
+preview_url: https://storage.googleapis.com/pyplots-images/plots/histogram-density/altair/plot.png
+preview_thumb: https://storage.googleapis.com/pyplots-images/plots/histogram-density/altair/plot_thumb.png
+preview_html: https://storage.googleapis.com/pyplots-images/plots/histogram-density/altair/plot.html
+quality_score: 93
+review:
+  strengths:
+  - Excellent bimodal distribution demonstrates density histogram features very effectively
+  - Clean visual design with proper color contrast between histogram bars and KDE
+    overlay
+  - Proper density normalization with clear axis labeling including units
+  - Interactive tooltips provide additional value for exploration
+  - Realistic cognitive psychology scenario with plausible reaction time values
+  weaknesses:
+  - Missing legend to identify the KDE line (yellow line has no label)
+  - Could use Altair's built-in transform_density() instead of scipy for more idiomatic
+    implementation