Add remaining ToCs to ToC lint (#56487)

Summary:
The lint was originally added in https://github.com/pytorch/pytorch/issues/54974, but at the time I didn't realize that these other Markdown files also each have a table of contents:

- `GLOSSARY.md`
- `torch/csrc/jit/OVERVIEW.md`
- `torch/csrc/jit/docs/serialization.md`
- `torch/fx/OVERVIEW.md`

This PR adds those files to the lint, and also changes the rule from using a fixed list of filenames to a `git grep` command that finds all Markdown files containing this magic comment:

```md

```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/56487

Test Plan: The "Lint / toc" job in GitHub Actions.

Reviewed By: janeyx99

Differential Revision: D27884885

Pulled By: samestep

fbshipit-source-id: 5462437502b17fba93abf5098e21754bf566a4fe

.github/workflows/lint.yml

@@ -128,7 +128,7 @@ jobs:
         run: |
           set -eux
           export PATH=~/.npm-global/bin:"$PATH"
-          for FILE in {CONTRIBUTING,README}.md; do
+          for FILE in $(git grep -Il '<!-- toc -->' -- '**.md'); do
             markdown-toc --bullets='-' -i "$FILE"
           done
       - name: Assert that regenerating the ToCs didn't change them

GLOSSARY.md

@@ -1,6 +1,7 @@
 # PyTorch Glossary
 
-- [PyTorch Glossary](#pytorch-glossary)
+<!-- toc -->
+
 - [Operation and Kernel](#operation-and-kernel)
   - [ATen](#aten)
   - [Operation](#operation)
@@ -19,6 +20,8 @@
   - [Tracing](#tracing)
   - [Scripting](#scripting)
 
+<!-- tocstop -->
+
 # Operation and Kernel
 
 ## ATen

torch/csrc/jit/OVERVIEW.md

@@ -14,8 +14,8 @@ Sections start with a reference to the source file where the code related to the
 
 ## Table of Contents
 
-- [JIT Technical Overview](#jit-technical-overview)
-  - [Table of Contents](#table-of-contents)
+<!-- toc -->
+
 - [Core Program Representation](#core-program-representation)
   - [Modules](#modules)
   - [Parameters](#parameters)
@@ -26,6 +26,7 @@ Sections start with a reference to the source file where the code related to the
   - [Block](#block)
     - [If](#if)
     - [Loops](#loops)
+    - [With](#with)
   - [Value](#value)
   - [Type](#type)
 - [Generating Programs](#generating-programs)
@@ -37,11 +38,12 @@ Sections start with a reference to the source file where the code related to the
   - [Lexer](#lexer)
   - [Tokens](#tokens)
   - [Parser](#parser)
-  - [Compiler](#compiler)
+  - [IR Emitter](#ir-emitter)
   - [SugaredValue](#sugaredvalue)
   - [Resolver](#resolver)
   - [Environment](#environment)
-  - [SSA Conversion](#convert_to_ssa)
+  - [Conversion To SSA](#conversion-to-ssa)
+  - [Exit Transform](#exit-transform)
   - [Python-Compiler Interaction](#python-compiler-interaction)
 - [Executing Programs](#executing-programs)
   - [Evaluation Semantics](#evaluation-semantics)
@@ -59,7 +61,7 @@ Sections start with a reference to the source file where the code related to the
     - [Aliasing and mutation in the PyTorch API](#aliasing-and-mutation-in-the-pytorch-api)
     - [Aliasing and mutation annotations in FunctionSchema](#aliasing-and-mutation-annotations-in-functionschema)
     - [Alias Analysis in the IR](#alias-analysis-in-the-ir)
-    - [Writing optimization passes with AliasDb](#writing-optimization-passes-with-aliasdb)
+    - [Writing optimization passes with `AliasDb`](#writing-optimization-passes-with-aliasdb)
 - [Profiling Programs](#profiling-programs)
 - [Saving Programs](#saving-programs)
 - [Testing Programs](#testing-programs)
@@ -67,6 +69,7 @@ Sections start with a reference to the source file where the code related to the
   - [Python Printer](#python-printer)
 - [Python Bindings](#python-bindings)
 
+<!-- tocstop -->
 
 # Core Program Representation
 

torch/csrc/jit/docs/serialization.md

@@ -3,6 +3,8 @@
 This document explains the TorchScript serialization format, and the anatomy
 of a call to `torch::jit::save()` or `torch::jit::load()`.
 
+<!-- toc -->
+
 - [Overview](#overview)
   - [Design Notes](#design-notes)
 - [`code/`: How code is serialized](#code-how-code-is-serialized)
@@ -10,14 +12,16 @@ of a call to `torch::jit::save()` or `torch::jit::load()`.
   - [Placing the source code in the archive](#placing-the-source-code-in-the-archive)
 - [How data is serialized](#how-data-is-serialized)
   - [`data.pkl`: How module object state is serialized](#datapkl-how-module-object-state-is-serialized)
-    - [`data/`: How tensors are serialized](#tensors-how-tensors-are-serialized)
+  - [`data/`: How tensors are serialized](#data-how-tensors-are-serialized)
 - [`constants.pkl`: Constants in code](#constantspkl-constants-in-code)
 - [`torch:jit::load()`](#torchjitload)
-  - [`__getstate__` and `__setstate__`](#getstate-and-setstate)
+- [`__getstate__` and `__setstate__`](#__getstate__-and-__setstate__)
 - [Appendix: `CompilationUnit` and code object ownership](#appendix-compilationunit-and-code-object-ownership)
   - [`CompilationUnit` ownership semantics](#compilationunit-ownership-semantics)
   - [Code object naming](#code-object-naming)
 
+<!-- tocstop -->
+
 ## Overview
 
 A serialized model (call it `model.pt`) is a ZIP archive containing many

torch/fx/OVERVIEW.md

@@ -4,21 +4,22 @@ FX is a toolkit for pass writers to facilitate Python-to-Python transformation o
 
 ## Table of Contents
 
-- [FX Technical Overview](#fx-technical-overview)
-  - [Table of Contents](#table-of-contents)
+<!-- toc -->
+
 - [Introduction](#introduction)
   - [Motivation](#motivation)
   - [Use Cases](#use-cases)
   - [Technical Details](#technical-details)
 - [Internal Structure](#internal-structure)
   - [Graph](#graph)
-  - [Graph Module](#graph-module)
+  - [GraphModule](#graphmodule)
 - [Symbolic Tracing](#symbolic-tracing)
-  - [About](#about)
   - [Tracer](#tracer)
   - [Proxy](#proxy)
-- [The FX IR](#ir)
-- [Transformation and Codegen](#codegen)
+- [The FX IR](#the-fx-ir)
+- [Transformation and Codegen](#transformation-and-codegen)
+
+<!-- tocstop -->
 
 # Introduction