[libcu++] Implement tuple protocol for integer_sequence

[davebayer]

This PR implements tuple protocol for integer_sequence (P1789R3) from C++26

[coderabbitai]

📝 Walkthrough

Summary by CodeRabbit

• New Features

  • cuda::std::integer_sequence now supports tuple protocol: tuple_size, tuple_element, and get() for mutable and const forms.
  • Improved compiler feature detection for structured-binding pack support.

• Tests

  • Added tests validating tuple-interface integration, structured bindings, and noexcept/get behavior.

• Chores

  • Updated library feature-test/versioning for integer_sequence and copyright years.

suggestion:

Walkthrough

Centralizes structured-bindings-pack detection, updates visit.cuh to use the new predicate, adds tuple-protocol specializations (tuple_size, tuple_element, get) for cuda::std::integer_sequence, bumps the integer_sequence feature-test macro, and adds tests validating tuple-like access and structured bindings.

Changes

Structured bindings pack support and tuple protocol

Layer / File(s)	Summary
Structured bindings pack feature detection libcudacxx/include/cuda/std/__cccl/dialect.h	Introduces _CCCL_HAS_STRUCTURED_BINDINGS_PACK() macro that detects pack support via __cpp_structured_bindings and compiler-version thresholds; explicitly disables it under NVCC.
Visit API refactoring cudax/include/cuda/experimental/__execution/visit.cuh	Replaces local __cpp_structured_bindings-based checks with _CCCL_HAS_STRUCTURED_BINDINGS_PACK() to select between pack-based and fixed-arity unrolling implementations in visit_t.
Tuple protocol for integer_sequence libcudacxx/include/cuda/std/__utility/integer_sequence.h, libcudacxx/include/cuda/std/version	Adds cuda::std::tuple_size, cuda::std::tuple_element (mutable and const), and cuda::std::get overload with bounds-check static_asserts; adds parallel std:: specializations; bumps __cccl_lib_integer_sequence to 202511L.
Tuple protocol integration tests libcudacxx/test/libcudacxx/std/utilities/intseq/intseq.binding/tuple_interface.pass.cpp	Adds a host/device constexpr test validating tuple_size_v, tuple_element_t, get<I> noexcept, and structured bindings via static_assert and runtime assert.

Suggested reviewers

• ericniebler
• elstehle
• NaderAlAwar

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (2)

libcudacxx/include/cuda/std/__utility/integer_sequence.h (2)

248-248: ⚡ Quick win

suggestion: Fix typo in comment.

"sequnence" → "sequence"

-// specialize cuda::std::tuple_size and cuda::std::tuple_element for cuda::std::integer_sequnence
+// specialize cuda::std::tuple_size and cuda::std::tuple_element for cuda::std::integer_sequence

---

280-280: ⚡ Quick win

suggestion: Fix typo in comment.

"sequnence" → "sequence"

-// tuple protocol for cuda::std::integer_sequnence
+// tuple protocol for cuda::std::integer_sequence

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 65b20831-4197-4671-a73f-1ca436bac9df

📥 Commits

Reviewing files that changed from the base of the PR and between 8623dab and d649c52.

📒 Files selected for processing (5)

• cudax/include/cuda/experimental/__execution/visit.cuh
• libcudacxx/include/cuda/std/__cccl/dialect.h
• libcudacxx/include/cuda/std/__utility/integer_sequence.h
• libcudacxx/include/cuda/std/version
• libcudacxx/test/libcudacxx/std/utilities/intseq/intseq.binding/tuple_interface.pass.cpp

[coderabbitai]

🧹 Nitpick comments (1)

libcudacxx/test/libcudacxx/std/utilities/intseq/intseq.binding/tuple_interface.pass.cpp (1)

59-64: ⚡ Quick win

suggestion: add a const auto [cv0, cv1, cv2, cv3] = seq4; case here. The current test only instantiates the non-const std::tuple_element path through structured bindings, so the new const specialization added in the header can still regress without this file noticing.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 36c3fd09-2d49-48bb-b9f1-60fba8c38625

📥 Commits

Reviewing files that changed from the base of the PR and between d649c52 and 6fdb61b.

📒 Files selected for processing (5)

• cudax/include/cuda/experimental/__execution/visit.cuh
• libcudacxx/include/cuda/std/__cccl/dialect.h
• libcudacxx/include/cuda/std/__utility/integer_sequence.h
• libcudacxx/include/cuda/std/version
• libcudacxx/test/libcudacxx/std/utilities/intseq/intseq.binding/tuple_interface.pass.cpp

[github-actions]

😬 CI Workflow Results

🟥 Finished in 1h 45m: Pass: 99%/159 | Total: 2d 03h | Max: 1h 44m | Hits: 69%/618936

See results here.