[Triton-MLIR] Enable libdevice for ptx backend when has external functions. (#848)

At the phase from ptx to cubin, check whether llvm::Module has external
functions. if has, link with libdevice at:
https://github.com/openai/triton/blob/triton-mlir/python/triton/language/libdevice.10.bc

python/tests/test_core.py

@@ -144,7 +144,7 @@ def _test_unary(dtype_x, expr, numpy_expr=None, device='cuda'):
     # triton result
     x_tri = to_triton(x, device=device, dst_type=dtype_x)
     z_tri = to_triton(np.empty_like(z_ref), device=device, dst_type=dtype_x)
-    kernel[(1, )](z_tri, x_tri, SIZE=SIZE, num_warps=4, extern_libs={"libdevice": "/usr/local/cuda/nvvm/libdevice/libdevice.10.bc"})
+    kernel[(1, )](z_tri, x_tri, SIZE=SIZE, num_warps=4)
     # compare
     np.testing.assert_allclose(z_ref, to_numpy(z_tri), rtol=0.01)