Native support for xAI Grok3

Model shorthand mapping related fixes
Comprehensive auto-mode related tests

.env.example

@@ -18,6 +18,9 @@ GEMINI_API_KEY=your_gemini_api_key_here
 # Get your OpenAI API key from: https://platform.openai.com/api-keys
 OPENAI_API_KEY=your_openai_api_key_here
 
+# Get your X.AI API key from: https://console.x.ai/
+XAI_API_KEY=your_xai_api_key_here
+
 # Option 2: Use OpenRouter for access to multiple models through one API
 # Get your OpenRouter API key from: https://openrouter.ai/
 # If using OpenRouter, comment out the native API keys above
@@ -68,15 +71,25 @@ DEFAULT_THINKING_MODE_THINKDEEP=high
 #   - flash                             (shorthand for gemini-2.5-flash-preview-05-20)
 #   - pro                               (shorthand for gemini-2.5-pro-preview-06-05)
 #
+# Supported X.AI GROK models:
+#   - grok-3          (131K context, advanced reasoning)
+#   - grok-3-fast     (131K context, higher performance but more expensive)
+#   - grok            (shorthand for grok-3)
+#   - grok3           (shorthand for grok-3)
+#   - grokfast        (shorthand for grok-3-fast)
+#
 # Examples:
 #   OPENAI_ALLOWED_MODELS=o3-mini,o4-mini,mini  # Only allow mini models (cost control)
 #   GOOGLE_ALLOWED_MODELS=flash                  # Only allow Flash (fast responses)
+#   XAI_ALLOWED_MODELS=grok-3                    # Only allow standard GROK (not fast variant)
 #   OPENAI_ALLOWED_MODELS=o4-mini                # Single model standardization
 #   GOOGLE_ALLOWED_MODELS=flash,pro              # Allow both Gemini models
+#   XAI_ALLOWED_MODELS=grok,grok-3-fast          # Allow both GROK variants
 #
 # Note: These restrictions apply even in 'auto' mode - Claude will only pick from allowed models
 # OPENAI_ALLOWED_MODELS=
 # GOOGLE_ALLOWED_MODELS=
+# XAI_ALLOWED_MODELS=
 
 # Optional: Custom model configuration file path
 # Override the default location of custom_models.json