feat: add recommendation on indexes in unique columns (#6553)

Co-authored-by: Jon Harrell <[email protected]>

content/700-optimize/300-recordings.mdx

@@ -45,6 +45,7 @@ When a recording session ends, Optimize generates recommendations such as:
 - [Using `@db.Char(n)`](/optimize/recommendations/avoid-char)
 - [Using `@db.VarChar(n)`](/optimize/recommendations/avoid-varchar)
 - [Using `timestamp(0)` or `timestamptz(0)`](/optimize/recommendations/avoid-timestamp-timestampz-0)
+- [Indexing on unique columns](/optimize/recommendations/indexing-on-unique-columns)
 - [Long-running transactions](/optimize/recommendations/long-running-transactions)
 - [Unnecessary indexes](/optimize/recommendations/unnecessary-indexes)
 

content/700-optimize/400-recommendations/1300-indexing-on-unique-columns.mdx

@@ -0,0 +1,17 @@
+---
+title: 'Indexing on unique columns'
+metaTitle: 'Optimize recommendations: Indexing on unique columns'
+metaDescription: "Learn about the recommendation provided by Optimize regarding indexing on uniquely constrained columns."
+tocDepth: 3
+toc: true
+---
+
+Optimize identifies redundant indexing on unique columns and provides recommendations for better database performance.
+
+### Why this is an issue
+
+Unique constraints inherently enforce uniqueness by generating an underlying index. Adding an additional index to the same column is unnecessary and can lead to extra overhead. 
+
+This redundancy increases write costs and slows down updates, as the database must synchronize multiple indexes.
+
+This guideline applies broadly to relational databases like PostgreSQL, MySQL, MariaDB, SQLite, and SQL Server, which automatically create indexes for unique constraints.