[INFRA] Import NVIDIA/CCCL upstream as optimization reference library

CCCL (CUDA C++ Core Libraries) provides:
- CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk)
- Thrust: high-level parallel algorithms (transform_reduce, sort, scan)
- libcudacxx: CUDA C++ standard library (atomics, barriers, memory)
- cudax: experimental features (memory resources, allocators)
- Tuning policies: per-SM hardware-specific algorithm parameters

Competition optimization vectors mapped to CCCL:
- Output TPS (83% weight): warp_reduce, block_reduce, device_topk
- Input TPS (14% weight): device_scan, block_load, prefetch
- Cache TPS (3% weight): prefix caching strategy patterns
- Memory (0.9 util): pooled/cached/buddy allocators

Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only)
License: Apache-2.0

cccl_upstream/libcudacxx/test/support/emplace_constructible.h

@@ -0,0 +1,86 @@
+#ifndef TEST_SUPPORT_EMPLACE_CONSTRUCTIBLE_H
+#define TEST_SUPPORT_EMPLACE_CONSTRUCTIBLE_H
+
+#include "test_macros.h"
+
+template <class T>
+struct EmplaceConstructible
+{
+  T value;
+  explicit EmplaceConstructible(T xvalue)
+      : value(xvalue)
+  {}
+  EmplaceConstructible(EmplaceConstructible const&) = delete;
+};
+
+template <class T>
+struct EmplaceConstructibleAndMoveInsertable
+{
+  int copied = 0;
+  T value;
+  explicit EmplaceConstructibleAndMoveInsertable(T xvalue)
+      : value(xvalue)
+  {}
+
+  EmplaceConstructibleAndMoveInsertable(EmplaceConstructibleAndMoveInsertable&& Other)
+      : copied(Other.copied + 1)
+      , value(std::move(Other.value))
+  {}
+};
+
+template <class T>
+struct EmplaceConstructibleAndMoveable
+{
+  int copied   = 0;
+  int assigned = 0;
+  T value;
+  explicit EmplaceConstructibleAndMoveable(T xvalue) noexcept
+      : value(xvalue)
+  {}
+
+  EmplaceConstructibleAndMoveable(EmplaceConstructibleAndMoveable&& Other) noexcept
+      : copied(Other.copied + 1)
+      , value(std::move(Other.value))
+  {}
+
+  EmplaceConstructibleAndMoveable& operator=(EmplaceConstructibleAndMoveable&& Other) noexcept
+  {
+    copied   = Other.copied;
+    assigned = Other.assigned + 1;
+    value    = std::move(Other.value);
+    return *this;
+  }
+};
+
+template <class T>
+struct EmplaceConstructibleMoveableAndAssignable
+{
+  int copied   = 0;
+  int assigned = 0;
+  T value;
+  explicit EmplaceConstructibleMoveableAndAssignable(T xvalue) noexcept
+      : value(xvalue)
+  {}
+
+  EmplaceConstructibleMoveableAndAssignable(EmplaceConstructibleMoveableAndAssignable&& Other) noexcept
+      : copied(Other.copied + 1)
+      , value(std::move(Other.value))
+  {}
+
+  EmplaceConstructibleMoveableAndAssignable& operator=(EmplaceConstructibleMoveableAndAssignable&& Other) noexcept
+  {
+    copied   = Other.copied;
+    assigned = Other.assigned + 1;
+    value    = std::move(Other.value);
+    return *this;
+  }
+
+  EmplaceConstructibleMoveableAndAssignable& operator=(T xvalue)
+  {
+    value = std::move(xvalue);
+    ++assigned;
+    return *this;
+  }
+};
+
+#endif // TEST_SUPPORT_EMPLACE_CONSTRUCTIBLE_H