fix(cuda-macros,codegen): keep all-generic library .oxart via #72 anchor

[dundysm]

Summary

Fixes #1365.

Depends on #1446. The backend stub calls oxide_artifacts::build_host_strong_anchor_object_for_target(_with_legacy_anchor), which #1446 adds in oxide-artifacts 0.2.2. The backend consumes oxide-artifacts from crates.io, so this PR builds once 0.2.2 is published and #1446 moves the requirement and relocks (the bump, publish, relock order from check-oxide-artifacts-parity.sh). After #1446 merges I will rebase this onto it and relock the new example with scripts/sync-example-locks.sh so it resolves 0.2.2 too.

An all-generic #[cuda_module] in a library that monomorphizes locally still embeds .oxart with an artifact anchor, but the macro skipped the #72 rlib keep-alive under the assumption that generics never produce an artifact in the defining crate. Nothing referenced the anchor, so the linker dropped the archive member and load_all_ptx_bundles_merged returned NoModules.

Adding one unused concrete kernel "fixed" it only because concrete kernels restored the keep-alive. This PR restores the handshake for generics without reverting #222 merge loading.

Root cause

Shape	Library .oxart?	Macro anchor ref?	Result
Concrete in lib (#72)	Yes	Yes	Works
Generics mono only in binary (#222)	Usually no	No (was skipped)	Works via merge
Generics mono in library, thin binary (#1365)	Yes	No (bug)	NoModules

Approach

Preserves the intentional design layers:

1. ModuleNotFound after I move a #[cuda_module] into a separate crate #72 archive handshake: keep referencing the package (or v2) artifact anchor from load_named.
2. Panic DriverError(500, "named symbol not found") when calling generic kernels from a separate crate #222 merge load: generic modules still call load_all_ptx_bundles_merged (no named-only regression).
3. No false undefined symbols: when the defining crate has merge markers but kernel_count == 0, the backend emits a strong anchor-only .oxlink stub (no fake empty PTX payload).

Macro (cuda_module_artifact_anchor_references)

• Remove the early return that skipped when every kernel is generic.
• Emit the same cfg-guarded black_box anchor refs for generic kernels via effective_cfg_attrs.
• Keep owner-filter / missing CARGO_PKG_* skips.

Backend (rustc-codegen-cuda)

• Extract cgus_require_ptx_bundle_merge so markers are scanned even without device code.
• When owner-selected, markers present, and there is no device codegen path (kernel_count == 0 / no device fns), emit a strong primary anchor-only object (v2 + weak legacy when owner filter is active), using the 0.2.2 writers from feat(oxide-artifacts): add strong anchor-only host objects, bump to 0.2.2 #1446.

Tests / example

• Macro unit: all-generic expansion keeps the merge loader, and the anchor reference builder (artifact_anchor_reference_tokens, split out of the env lookup so it can be tested without cargo's per-crate env) emits the keep-alive for generic kernels; cfg-gated generic gates the ref.
• oxide-artifacts strong anchor-only object tests now live in feat(oxide-artifacts): add strong anchor-only host objects, bump to 0.2.2 #1446.
• New generic_mono_in_lib two-crate example under cuda-oxide/crates/rustc-codegen-cuda/examples/ matching the issue (lib mono + thin bin).
• Existing cuda_module_in_lib / cross_crate_embedded shapes remain the concrete and Panic DriverError(500, "named symbol not found") when calling generic kernels from a separate crate #222 regressions.

Orthogonal to #1367 / #1166 (finalizer expected-kernel inventory).

Test plan

• cargo test -p cuda-macros (118 unit tests, including the two generic anchor tests)
• cargo clippy -p cuda-macros --all-targets -- -D warnings
• cargo fmt --check for the SIMT workspace, the backend, and the new example
• rustc-codegen-cuda: cargo check, cargo clippy --all-targets -- -D warnings, and cargo test against the feat(oxide-artifacts): add strong anchor-only host objects, bump to 0.2.2 #1446 crate via a local [patch.crates-io]
• scripts/sync-example-locks.sh --check and every cuda-oxide/scripts/check-*.sh, including the cargo-deny example license policy
• cargo oxide run generic_mono_in_lib and --verify-bundles
• cargo oxide run cuda_module_in_lib and cross_crate_embedded

Reviewers

cc @nihalpasham @xavierforge (prior #72 / from-anchor / embedding owners; CODEOWNERS currently declares no path owners in-tree)

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[dundysm]

split the oxide-artifacts part out into #1446 (new strong anchor-only writers, bump to 0.2.2). the backend reads oxide-artifacts from crates.io, so this one now builds once 0.2.2 is published and relocked; i'll rebase it on top after that lands. also squashed here: the example moved under cuda-oxide/crates/rustc-codegen-cuda/examples, rustfmt, and the generic anchor tests now go through a small token builder so they pass under plain cargo test.