fix: add index to navigation and improve landing page design

Author: hallelx2

docs/docs.json

@@ -39,6 +39,7 @@
   ],
   "navigation": {
     "pages": [
+      "index",
       "getting-started/installation",
       "getting-started/quickstart",
       "getting-started/concepts",

docs/index.md

@@ -1,19 +1,27 @@
+---
+title: Home
+description: SwarmTorch - 120 Metaheuristic Optimization Algorithms for PyTorch
+---
+
 # SwarmTorch 🐝🔥
 
-<Callout type="info">
-  SwarmTorch brings 120 metaheuristic optimization algorithms to PyTorch — 60 model-training optimizers and 60 hyperparameter-tuning searchers.
-</Callout>
+<Info>
+SwarmTorch brings 120 metaheuristic optimization algorithms to PyTorch — 60 model-training optimizers and 60 hyperparameter-tuning searchers.
+</Info>
 
 SwarmTorch is a high-performance, academic-grade library that enables **gradient-free neural network training** and **intelligent hyperparameter optimization** using nature-inspired metaheuristic algorithms.
 
 ## Why SwarmTorch?
 
-Traditional deep learning relies on gradient-based optimization (Adam, SGD, etc.). SwarmTorch complements these with:
+Traditional deep learning relies on gradient-based optimization (Adam, SGD). SwarmTorch complements these with:
 
-- **Gradient-Free Training**: Optimize weights where gradients don't exist or are unreliable
-- **Hyperparameter Optimization**: Replace grid/random search with intelligent exploration
-- **60+ Algorithms**: From PSO and GWO to DE, GA, and hybrid methods
-- **PyTorch Native**: Drop-in replacement for standard optimizers
+| Scenario | Gradient (Adam/SGD) | SwarmTorch |
+|----------|-------------------|------------|
+| Standard classification | ✅ Best choice | Overkill |
+| Non-differentiable loss | ❌ Can't use | ✅ Perfect fit |
+| Discrete optimization | ❌ Can't use | ✅ Perfect fit |
+| Multi-modal landscapes | Gets stuck | Explores well |
+| Hyperparameter tuning | Manual search | Automated |
 
 ## Key Features
 
@@ -24,7 +32,7 @@ Traditional deep learning relies on gradient-based optimization (Adam, SGD, etc.
   <Card title="60 HPO Searchers" icon="magnifying-glass">
     Automatically find optimal hyperparameters using nature-inspired search.
   </Card>
-  <Card title="PyTorch Compatible" icon="brand-elixir">
+  <Card title="PyTorch Native" icon="brand-elixir">
     Uses the standard `torch.optim.Optimizer` interface. Easy to integrate.
   </Card>
   <Card title="Research-Ready" icon="chart-line">
@@ -54,8 +62,8 @@ for _ in range(100):
 
 ## Algorithm Categories
 
-| Category | # Algorithms | Examples |
-|----------|-------------|----------|
+| Category | # | Examples |
+|----------|---|----------|
 | **Swarm Intelligence** | 32 | PSO, GWO, WOA, HHO, SSA |
 | **Evolutionary** | 8 | DE, GA, CEM, PBIL |
 | **Physics-Based** | 3 | SA, GSA, FPA |
@@ -70,11 +78,12 @@ pip install swarmtorch
 ```
 
 For benchmarking dependencies:
+
 ```bash
 pip install swarmtorch[benchmarks]
 ```
 
-## Next Steps
+## Get Started
 
 <CardGroup cols={2}>
   <Card title="Quickstart" icon="rocket" href="/getting-started/quickstart">
@@ -86,7 +95,13 @@ pip install swarmtorch[benchmarks]
   <Card title="Hyperparameter Tuning" icon="magnifying-glass" href="/guides/hyperparameter-tuning">
     Discover how to optimize model hyperparameters automatically.
   </Card>
+  <Card title="Choosing an Algorithm" icon="sparkles" href="/guides/choosing-algorithm">
+    Find the right algorithm for your problem.
+  </Card>
   <Card title="API Reference" icon="code" href="/api-reference/base">
     Explore the full API documentation.
   </Card>
+  <Card title="Benchmarks" icon="chart-bar" href="/benchmarks/benchmarks">
+    See where swarm algorithms excel vs gradient methods.
+  </Card>
 </CardGroup>