Fix forward AD tutorial directory and add to index (#1758)

Author: soulitzer

index.rst

@@ -387,6 +387,13 @@ Welcome to PyTorch Tutorials
    :link: intermediate/custom_function_conv_bn_tutorial.html
    :tags: Extending-PyTorch,Frontend-APIs
 
+.. customcarditem::
+   :header: Forward-mode Automatic Differentiation
+   :card_description: Learn how to use forward-mode automatic differentiation
+   :image: _static/img/thumbnails/cropped/generic-pytorch-logo.PNG
+   :link: intermediate/forward_ad_usage.html
+   :tags: Frontend-APIs
+
 .. Model Optimization
 
 .. customcarditem::
@@ -746,6 +753,7 @@ Additional Resources
    :caption: Frontend APIs
 
    intermediate/memory_format_tutorial
+   intermediate/forward_ad_usage
    advanced/cpp_frontend
    advanced/torch-script-parallelism
    advanced/cpp_autograd

forward_ad_usage.py renamed to intermediate_source/forward_ad_usage.py

@@ -1,11 +1,17 @@
 # -*- coding: utf-8 -*-
 """
-Forward-mode Automatic Differentiation
-======================================
+Forward-mode Automatic Differentiation (Beta)
+=============================================
 
 This tutorial demonstrates how to use forward-mode AD to compute
 directional derivatives (or equivalently, Jacobian-vector products).
 
+The tutorial below uses some APIs only available in versions >= 1.11
+(or nightly builds).
+
+Also note that forward-mode AD is currently in beta. The API is
+subject to change and operator coverage is still incomplete.
+
 Basic Usage
 --------------------------------------------------------------------
 Unlike reverse-mode AD, forward-mode AD computes gradients eagerly