⚡ feat: add Phase 5 CUDA Graph optimization for chunked prefill

Implement extended CUDA Graph coverage for CPU offload path:
- Add graphed_layers.py with N+2 graph architecture (EmbedGraph, FirstGraph, InterGraphs, LastGraph)
- Support both prefill (seq_len=chunk_size) and decode (seq_len=1) graph modes
- Extend graph coverage to ~70-80% including qkv_proj, rotary, o_proj
- Only attention core remains in eager mode for dynamic offload

Performance: Prefill throughput improved ~5.6% (3782 -> 3995 tok/s at 32K)

Also adds:
- --enforce-eager flag to bench_offload.py for comparison
- Offload mode constraint documentation in CLAUDE.md

Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)

Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>

bench_offload.py

@@ -69,6 +69,7 @@ def main():
     parser.add_argument("--max-len", type=int, default=32*1024, help="Max model length (default: 32K)")
     parser.add_argument("--bench-decode", action="store_true", help="Run decode benchmark (default: prefill only)")
     parser.add_argument("--bench-all", action="store_true", help="Run both prefill and decode benchmarks")
+    parser.add_argument("--enforce-eager", action="store_true", help="Disable CUDA Graphs (use eager mode)")
     args = parser.parse_args()
 
     path = os.path.expanduser(args.model)
@@ -89,7 +90,7 @@ def main():
 
     llm = LLM(
         path,
-        enforce_eager=False,
+        enforce_eager=args.enforce_eager,
         max_model_len=max_len,
         max_num_batched_tokens=max_len,
         enable_cpu_offload=True,