Add .apr/.ald display demo example

- apr_ald_display.rs demonstrates loading and displaying
- Shows ModelCard from .apr with metrics, params, layers
- Shows DataCard from .ald with tensor info
- Both direct loading and extension trait usage

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

Author: noahgift

crates/presentar/examples/apr_ald_display.rs

@@ -0,0 +1,109 @@
+//! Demo: Display .apr model and .ald dataset files
+//!
+//! Run with: cargo run --example apr_ald_display
+
+use presentar_widgets::{load_ald_as_card, load_apr_as_card, AldDatasetExt, AprModelExt};
+use presentar_yaml::formats::{AldDataset, AprModel, ModelLayer, Tensor};
+
+fn main() {
+    println!("=== Presentar .apr/.ald Display Demo ===\n");
+
+    // Create a sample .apr model
+    let model = create_sample_model();
+    let model_bytes = model.save();
+    println!("Created sample .apr model ({} bytes)\n", model_bytes.len());
+
+    // Load and display as ModelCard
+    let model_card = load_apr_as_card(&model_bytes).expect("load apr");
+    println!("ModelCard from .apr:");
+    println!("  Name: {}", model_card.get_name());
+    println!("  Version: {}", model_card.get_version());
+    println!("  Framework: {:?}", model_card.get_framework());
+    println!("  Parameters: {:?}", model_card.get_parameters());
+    println!("  Status: {:?}", model_card.get_status());
+    println!("  Tags: {:?}", model_card.get_tags());
+    println!();
+
+    // Create a sample .ald dataset
+    let dataset = create_sample_dataset();
+    let dataset_bytes = dataset.save();
+    println!("Created sample .ald dataset ({} bytes)\n", dataset_bytes.len());
+
+    // Load and display as DataCard
+    let data_card = load_ald_as_card(&dataset_bytes, "mnist_train").expect("load ald");
+    println!("DataCard from .ald:");
+    println!("  Name: {}", data_card.get_name());
+    println!("  Columns: {}", data_card.column_count());
+    println!("  Description: {:?}", data_card.get_description());
+    println!("  Tags: {:?}", data_card.get_tags());
+    println!();
+
+    // Using extension traits directly
+    println!("=== Using Extension Traits ===\n");
+
+    let card = model.to_model_card();
+    println!("Direct conversion: {}", card.get_name());
+
+    let card = dataset.to_data_card("custom_name");
+    println!("Direct conversion: {}", card.get_name());
+
+    println!("\nDone!");
+}
+
+fn create_sample_model() -> AprModel {
+    let mut model = AprModel::new("MLP");
+
+    // Add layers
+    model.layers.push(ModelLayer {
+        layer_type: "dense".to_string(),
+        parameters: vec![
+            Tensor::from_f32("weights", vec![784, 256], &vec![0.0; 784 * 256]),
+            Tensor::from_f32("bias", vec![256], &vec![0.0; 256]),
+        ],
+    });
+
+    model.layers.push(ModelLayer {
+        layer_type: "relu".to_string(),
+        parameters: vec![],
+    });
+
+    model.layers.push(ModelLayer {
+        layer_type: "dense".to_string(),
+        parameters: vec![
+            Tensor::from_f32("weights", vec![256, 10], &vec![0.0; 256 * 10]),
+            Tensor::from_f32("bias", vec![10], &vec![0.0; 10]),
+        ],
+    });
+
+    // Add metadata
+    model
+        .metadata
+        .insert("accuracy".to_string(), "0.98".to_string());
+    model
+        .metadata
+        .insert("task".to_string(), "classification".to_string());
+    model
+        .metadata
+        .insert("dataset".to_string(), "MNIST".to_string());
+    model
+        .metadata
+        .insert("author".to_string(), "PAIML".to_string());
+
+    model
+}
+
+fn create_sample_dataset() -> AldDataset {
+    let mut dataset = AldDataset::new();
+
+    // Add training images (simulated)
+    dataset.add_tensor(Tensor::from_f32(
+        "images",
+        vec![60000, 28, 28],
+        &vec![0.0; 60000 * 28 * 28],
+    ));
+
+    // Add labels
+    dataset.add_tensor(Tensor::from_f32("labels", vec![60000], &vec![0.0; 60000]));
+
+    dataset
+}