fix(workflow,kernel): split plan stage 'produces' (slot name) from 'kind' (artifact kind)

ExecutionPlanCompiler treated the plan's single 'produces' string as a
registry kind id, while the architect prompt described it as a unique
artifact id referenced by needs/edges — the model followed the prompt,
invented descriptive ids (files_created), and every freestyle plan
failed to compile. Collapsing both onto one string is also unsound:
two stages producing the same kind would collide on slot name and make
artifact_validated edge conditions ambiguous.

Plan stages now carry an optional 'kind' selecting the registered kind
(mirroring TOML's produces = { name, kind }); 'produces' stays the
unique slot name. Missing 'kind' falls back to the old produces-as-kind
reading so existing plans still compile. Vocabulary entry, architect
prompt, and execution_plan schemas updated to match.

core/kernel/src/main/kotlin/com/correx/core/kernel/orchestration/ContextFeedback.kt

@@ -54,7 +54,9 @@ fun buildArtifactKindVocabularyEntry(kinds: Collection<ArtifactKind>): ContextEn
         if (k.llmEmitted) "${k.id} (llm-emitted)" else k.id
     }
     val content = "## Available artifact kinds\n" +
-        "Every stage's \"produces\" MUST be exactly one of: $listing\n" +
+        "Every stage's \"kind\" MUST be exactly one of: $listing\n" +
+        "Stages that write or edit files use \"file_written\"; stages that run commands use " +
+        "\"process_result\".\n" +
         "Do not invent kinds — a plan referencing an unknown kind fails to compile."
     return ContextEntry(
         id = ContextEntryId(UUID.randomUUID().toString()),