Add section page for Rationale docs.

[richard-uhler]

Summary

With a brief overview and summary of each of the Rationale docs.

[rriddle]

Thanks!

mlir/docs/Rationale/_index.md

7

Would the - description work better as a sub bullet? How is this displayed on the website?

[jpienaar]

Thanks

mlir/docs/Rationale/_index.md

30

Nit (and this could be in doc too/not due to you) but s/'Const'/'const'/ given C keyword

[richard-uhler]

Switch to using markdown's definition list.

[richard-uhler]

Switch from Const to const.

[richard-uhler]

mlir/docs/Rationale/_index.md

7

I tried some variations. I like definition list best, so I switched to that. Screenshot attached:

.

[rriddle]

mlir/docs/Rationale/_index.md

7

LG, thanks.

mlir/docs/Rationale/_index.md

+# Rationale
+
+This section contains a collection of documents describing the motivation and
+rationale for some of the design decisions behind MLIR.
+
+[MLIR: Incremental Application to Graph Algorithms in ML Frameworks](MLIRForGraphAlgorithms.md)
+:   A discussion of how the adoption of MLIR can be taken in incremental steps,
+    with each step providing tangible benefits along the way. Refutes the idea
+    that full adoption of MLIR is required before we can reap the benefits of
+    MLIR.
+
+[MLIR Rationale](Rationale.md)
+:   Introduces the motivation for MLIR and captures design discussions and
+    decisions made for various core features of MLIR.
+
+[Generic DAG Rewriter Infrastructure Rationale](RationaleGenericDAGRewriter.md)
+:   Details the rationale behind a general DAG-to-DAG rewrite infrastructure for
+    MLIR.
+
+[Linalg Dialect Rationale: The Case for Compiler-Friendly Custom Operations](RationaleLinalgDialect.md)
+:   Describes the key design principles that led to the existing implementation
+    of Linalg and lessons learned along the way.
+
+[MLIR: The case for a simplified polyhedral form](RationaleSimplifiedPolyhedralForm.md)
+:   An early design proposal exploring the tradeoffs of using a simplified form
+    for polyhedral compiler techniques in MLIR instead of the traditional
+    polyhedral schedule list form.
+
+[Usage of 'Const' in MLIR, for core IR types](UsageOfConst.md)
+:   Explains the rationale for eschewing the use of `const` entirely for the
+    core IR types in MLIR.