fix(kernel+infra): tool approval gate and executor cleanup

- SessionOrchestrator: wire T2/T3/T4 approval gate before tool execution;
  emit OrchestrationPausedEvent/ApprovalRequestedEvent, await decision,
  return rejection as ERROR context entry so LLM sees the denial;
  propagate tool ERROR entries as StageExecutionResult.Failure
- SandboxedToolExecutor: remove dead code and simplify
- InfrastructureModule: minor wiring cleanup
- LlamaCppInferenceProvider / build.gradle: related build fixes

infrastructure/inference/llama_cpp/build.gradle

@@ -17,8 +17,6 @@ dependencies {
     implementation project(':core:context')
     implementation project(':infrastructure:inference:commons')
 
-    implementation("com.fasterxml.jackson.core:jackson-databind:2.17.0")
-    implementation("com.fasterxml.jackson.module:jackson-module-kotlin:2.17.0")
     implementation "io.ktor:ktor-client-core:$ktor_version"
     implementation "io.ktor:ktor-client-cio:$ktor_version"
     implementation "io.ktor:ktor-client-content-negotiation:$ktor_version"

infrastructure/inference/llama_cpp/src/main/kotlin/com/correx/infrastructure/inference/llama/cpp/LlamaCppInferenceProvider.kt

@@ -25,7 +25,9 @@ import io.ktor.client.request.setBody
 import io.ktor.http.ContentType
 import io.ktor.http.contentType
 import io.ktor.serialization.kotlinx.json.json
+import kotlinx.serialization.encodeToString
 import kotlinx.serialization.json.Json
+import org.slf4j.LoggerFactory
 
 private const val DEFAULT_REQUEST_TIMEOUT_MS = 60_000L
 
@@ -42,7 +44,23 @@ private fun defaultHttpClient(): HttpClient = HttpClient(CIO) {
     install(HttpTimeout) {
         requestTimeoutMillis = DEFAULT_REQUEST_TIMEOUT_MS
     }
+    install(ContentNegotiation) {
+        json(
+            Json {
+                ignoreUnknownKeys = true
+                explicitNulls = false
+                encodeDefaults = true
+            },
+        )
+    }
 }
+private val json = Json {
+    ignoreUnknownKeys = true
+    explicitNulls = false
+    encodeDefaults = true
+}
+
+private val log = LoggerFactory.getLogger(LlamaCppInferenceProvider::class.java)
 
 @Suppress("TooGenericExceptionCaught", "MagicNumber")
 class LlamaCppInferenceProvider(
@@ -80,12 +98,17 @@ class LlamaCppInferenceProvider(
             tools = request.tools.takeIf { it.isNotEmpty() },
         )
 
+        val encoded = json.encodeToString(body)
+        log.debug("sending request to llm: {}", encoded)
+
         val response = httpClient.post("$baseUrl/v1/chat/completions") {
             contentType(ContentType.Application.Json)
             accept(ContentType.Application.Json)
-            setBody(body)
+            setBody(encoded)
         }.body<ChatCompletionResponse>()
 
+        log.debug("got response from llm: {}", response)
+
         val message = response.choices.first().message
         val finishReason = when (response.choices.first().finishReason.lowercase()) {
             "length" -> FinishReason.Length