perf(runtime): Hermes Phase 1-3 — prompt caching + parallel tools + smart retry

Phase 1: Anthropic prompt caching
- Add cache_control ephemeral on system prompt blocks
- Track cache_creation/cache_read tokens in CompletionResponse + StreamChunk

Phase 2A: Parallel tool execution
- Add ToolConcurrency enum (ReadOnly/Exclusive/Interactive)
- JoinSet + Semaphore(3) for bounded parallel tool calls
- 7 tools annotated with correct concurrency level
- AtomicU32 for lock-free failure tracking in ToolErrorMiddleware

Phase 2B: Tool output pruning
- prune_tool_outputs() trims old ToolResult > 2000 chars to 500 chars
- Integrated into CompactionMiddleware before token estimation

Phase 3: Error classification + smart retry
- LlmErrorKind + ClassifiedLlmError for structured error mapping
- RetryDriver decorator with jittered exponential backoff
- Kernel wraps all LLM calls with RetryDriver
- CONTEXT_OVERFLOW recovery triggers emergency compaction in loop_runner

crates/zclaw-runtime/src/stream.rs

@@ -24,6 +24,10 @@ pub enum StreamChunk {
         input_tokens: u32,
         output_tokens: u32,
         stop_reason: String,
+        #[serde(default)]
+        cache_creation_input_tokens: Option<u32>,
+        #[serde(default)]
+        cache_read_input_tokens: Option<u32>,
     },
     /// Error occurred
     Error { message: String },