fix: workflow inference chain — surface llama-server errors, user-turn-last, bundle-relative prompts

- LlamaCppInferenceProvider: check HTTP status and surface the real error body instead of masking it as a ChatCompletionResponse deserialization failure
- PromptRenderer: render the live conversation layer (L1) last so the user turn is the final message (fixes Qwen 'No user query found' 500 when L3 recall is present)
- TomlWorkflowLoader: resolve stage prompts relative to the workflow file's own dir (bundle), CWD-independent; config-dir fallback for globals
- SessionOrchestrator: hard-fail a stage whose declared prompt can't resolve (was a silent user-less request)
- move healthcheck prompts next to the example workflow (examples/workflows/prompts/)

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

@@ -222,7 +222,7 @@ abstract class SessionOrchestrator(
             } ?: emptyList()
         val promptEntries = stageConfig.metadata["prompt"]
             ?.let { path ->
-                runCatching { promptResolver.resolve(path) }
+                val resolvedText = runCatching { promptResolver.resolve(path) }
                     .onFailure {
                         log.error(
                             "[SessionOrchestrator] stage=${stageId.value}: " +
@@ -230,9 +230,11 @@ abstract class SessionOrchestrator(
                         )
                     }
                     .getOrNull()
-            }
-            ?.takeIf { it.isNotBlank() }
-            ?.let { text ->
+                val text = resolvedText?.takeIf { it.isNotBlank() }
+                    ?: error(
+                        "[SessionOrchestrator] stage=${stageId.value}: " +
+                                "declared prompt '$path' could not be resolved",
+                    )
                 listOf(
                     ContextEntry(
                         id = ContextEntryId(UUID.randomUUID().toString()),