add note about how you dont need custom checkpointer (#117)

I've noticed a couple customers have gotten confused when deploying LG
agents to LGP, because they configure custom checkpointers and stores.
But this is not supported with LGP, because these are provided out of
the box. This docs change adds a Note on step 3 of getting started which
explains this.

I don't think this will stop many cases of this confusion, so long term
maybe we should think about a simpler way to show "getting started on
LGP". This tutorial is pretty long to begin with....

---------

Co-authored-by: Kathryn May <[email protected]>

Author: samecrowder

src/langgraph-platform/langgraph-basics/3-add-memory.mdx

@@ -15,6 +15,12 @@ LangGraph solves this problem through **persistent checkpointing**. If you provi
 
 We will see later that **checkpointing** is _much_ more powerful than simple chat memory - it lets you save and resume complex state at any time for error recovery, human-in-the-loop workflows, time travel interactions, and more. But first, let's add checkpointing to enable multi-turn conversations.
 
+<Note>
+If you're [deploying your graph to LangGraph Platform](/langgraph-platform/deployment-options), you do not need to set up a checkpointer manually — checkpointing and memory are handled for you automatically.
+
+This tutorial focuses on local development, where you do need to configure a checkpointer. When you're ready to deploy, remove this checkpointer configuration.
+</Note>
+
 ## 1. Create a `MemorySaver` checkpointer
 
 Create a `MemorySaver` checkpointer: