[clang] [llvm] AMDGPU: Move enqueued block handling into clang (PR #128519)

[Matt Arsenault via cfe-commits]

arsenm wrote:

* **#128520** <a href="https://app.graphite.dev/github/pr/llvm/llvm-project/128520?utm_source=stack-comment-icon" target="_blank"><img src="https://static.graphite.dev/graphite-32x32-black.png" alt="Graphite" width="10px" height="10px"/></a>
* **#128519** <a href="https://app.graphite.dev/github/pr/llvm/llvm-project/128519?utm_source=stack-comment-icon" target="_blank"><img src="https://static.graphite.dev/graphite-32x32-black.png" alt="Graphite" width="10px" height="10px"/></a> 👈 <a href="https://app.graphite.dev/github/pr/llvm/llvm-project/128519?utm_source=stack-comment-view-in-graphite" target="_blank">(View in Graphite)</a>
* `main`


<h2></h2>

This stack of pull requests is managed by <a href="https://graphite.dev?utm-source=stack-comment"><b>Graphite</b></a>. Learn more about <a href="https://stacking.dev/?utm_source=stack-comment">stacking</a>.
<!-- Current dependencies on/for this PR: -->

https://github.com/llvm/llvm-project/pull/128519