Debug Counters Implementation - Clean History

Major Features:
- Debug counter infrastructure for Refill Stage tracking
- Free Pipeline counters (ss_local, ss_remote, tls_sll)
- Diagnostic counters for early return analysis
- Unified larson.sh benchmark runner with profiles
- Phase 6-3 regression analysis documentation

Bug Fixes:
- Fix SuperSlab disabled by default (HAKMEM_TINY_USE_SUPERSLAB)
- Fix profile variable naming consistency
- Add .gitignore patterns for large files

Performance:
- Phase 6-3: 4.79 M ops/s (has OOM risk)
- With SuperSlab: 3.13 M ops/s (+19% improvement)

This is a clean repository without large log files.

🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>

core/hakmem_tiny_remote_target.h

@@ -0,0 +1,25 @@
+#ifndef HAKMEM_TINY_REMOTE_TARGET_H
+#define HAKMEM_TINY_REMOTE_TARGET_H
+
+#include <stdatomic.h>
+#include <stdint.h>
+
+// Forward declaration
+typedef struct TinySlab TinySlab;
+
+#define TINY_NUM_CLASSES 8
+
+// Targeted remote-drain queue: Slabs with high remote free counts
+// Background thread can drain these without blocking allocation hot path
+extern int g_bg_remote_enable;
+extern _Atomic uintptr_t g_remote_target_head[TINY_NUM_CLASSES];
+extern _Atomic uint32_t g_remote_target_len[TINY_NUM_CLASSES];
+extern int g_bg_remote_batch;
+
+// Enqueue a slab for targeted remote drain (lock-free Treiber stack)
+void remote_target_enqueue(int class_idx, TinySlab* slab);
+
+// Dequeue a slab from remote drain queue (returns NULL if empty)
+TinySlab* remote_target_pop(int class_idx);
+
+#endif // HAKMEM_TINY_REMOTE_TARGET_H