[Qiong Wu (qiowu) (DingmaomaoBJTU)]

Offline DML CG Output IR can fail with 0x80070032 when WinMLRuntimeSession sends an already compiled artifact through the compiler again, even though CLI compilation and reload validation succeeded.

Detect Output IR internally by parsing the top-level module interface attribute through the MLIR C API. A cgc_output interface loads directly into the Runtime pipeline; Input IR and ONNX retain the existing compilation path. Both text and bytecode artifacts are supported, without scanning filenames/comments or adding a public API flag. Classification happens before loading the Runtime model to avoid leaking native handles if parsing fails.

Validation:

• Native RTX 4080 checks with Windows ML Runtime 2.8.183.dev0 passed for both text and bytecode DML Output IR. Forcing the previous recompile path reproduces 0x80070032; automatic detection loads and runs the same artifact with exact expected outputs.
• Runtime-session and classification unit regressions: 96 passed.
• Ruff check passed.

Rebased on upstream main d7055f9. The original RTX 5090 D benchmark roster has not been rerun. No caller/API changes are required for Output IR detection.