Minor: Add comment explaining rationale to avoid dependencies on functions (#20667)

## Which issue does this PR close?

<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax. For example
`Closes #123` indicates that this PR will close issue #123.
-->

- Closes #.

## Rationale for this change

Somewhat frequently PRs try and add a dependency (both codewise and
logically) to datafusion-functions in the datafusion-optimizer crate

This then gets flagged during code review, which is somewhat frustrating
probably for the submitter as they didn't know. This most recently
happened here
https://github.com/apache/datafusion/pull/20610/changes#r2877698578

Thus let's try and document that attempt in comments

## What changes are included in this PR?

Add comments to the relevant Cargo.toml file

## Are these changes tested?

N/A
## Are there any user-facing changes?

No this is a development process change only

Author: alamb

datafusion/optimizer/Cargo.toml

@@ -43,6 +43,14 @@ name = "datafusion_optimizer"
 [features]
 recursive_protection = ["dep:recursive"]
 
+# Note -- please DO NOT add a dependency here to any of the datafusion-functions
+# crates. While it is tempting to try and add an optimizer pass that uses
+# datafusion-functions Doing so makes it harder for downstream crates to
+# provide their own function library and smaller install footprint.
+#
+# If you want to add special handling for a specific function, use the methods
+# on the ScalarUDFImpl or AggregateUDFImpl traits (or add a new method to those
+# traits).
 [dependencies]
 arrow = { workspace = true }
 chrono = { workspace = true }