chore(release): t27c 0.5.2 -- the first release the swarm makes #7965

Why

Owner, 2026-10-08 (translated): "then any release or multi-step job is done by the swarm, and nobody needs an open laptop -- YES, bring it to release". Asked again on 2026-10-09 whether to take it to a real t27c release, the owner answered (translated): "Yes: a rehearsal, then the release."

t27c 0.5.1 was released by hand (2026-10-08 11:42Z). The swarm has only rehearsed, on 0.5.0 (#7700), and that run correctly refused at tag-absent. A green rehearsal, and then a release made by the swarm, need a version that the manifests carry and no tag has. This issue is the bump to 0.5.2.

bootstrap/ changed since t27c-v0.5.1:

• t27core: use between modules, byte-identical to gen-c's splice (Closes self-host P1 slice 6: use between modules, byte-identical to gen-c's splice #7935) (t27core: use between modules, byte-identical to gen-c's splice (Closes #7935) #7950)
• t27core: local arrays, separators in dimensions, numeric ** refused (Closes self-host P1 slice 5: local arrays, separators in dimensions, numeric ** pinned as refused #7929) (t27core: local arrays, separators in dimensions, numeric ** refused (Closes #7929) #7931)
• fixtures(core_selfhost/genc): build the _ => arm again, as gen-c now lowers it as the catch-all (Closes core_selfhost genc/switch_exhaustive.t27: build the _ => arm again, now that gen-c lowers _ as the catch-all (#7456) #7879) (fixtures(core_selfhost/genc): build the _ => arm again, as gen-c now lowers it as the catch-all (Closes #7879) #7880)
• ci: test only the specs a change can affect -- t27c use-edges feeds specs/ci/affected.t27 (Closes ci: wire specs/ci/affected.t27 -- t27c use-edges + test only the affected specs #7565) (ci: test only the specs a change can affect -- t27c use-edges feeds specs/ci/affected.t27 (Closes #7565) #7566)
• fix(gen-c): a local nested array is declared as a C matrix (Closes t27c gen-c: local nested array declared as [16]u8 m[16] #7441) (fix(gen-c): a local nested array is declared as a C matrix (Closes #7441) #7495)
• t27core: for over one range and switch as a value, byte-identical to gen-c (Closes t27core: for over one range and switch as a value, byte-identical to gen-c #7429) (t27core: for over one range and switch as a value, byte-identical to gen-c (Closes #7429) #7436)
• t27c gen-c: size_t range counter, refuse an integer switch without else (Closes self-host: gen-c declares a range-for counter as int, not size_t #7307, Closes self-host: gen-c lowers a switch with no else arm to 0 for the missing case #7308) (t27c gen-c: size_t range counter, refuse an integer switch without else (Closes #7307, Closes #7308) #7331)

Acceptance criteria

• bootstrap/Cargo.toml, .zenodo.json and the t27c entry of Cargo.lock read 0.5.2, and cargo metadata --locked is consistent.
• No tag t27c-v0.5.2 exists when this merges. The tag is the swarm's job (card release-t27c), not this PR's.