Docling is slow for OCR using RapidOcr

### Bug
Getting a lot of error warnings while doing OCR with Rapid OCR and takes 40 secs to execute with GPU

### Steps to reproduce
`WARNING:OrtInferSession:DmlExecutionProvider is not in available providers (['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']). Use TensorrtExecutionProvider inference by default.
2025-03-11 20:39:23,411 - OrtInferSession - INFO: If you want to use DirectML acceleration, you must do:
INFO:OrtInferSession:If you want to use DirectML acceleration, you must do:
2025-03-11 20:39:23,411 - OrtInferSession - INFO: First, uninstall all onnxruntime pakcages in current environment.
INFO:OrtInferSession:First, uninstall all onnxruntime pakcages in current environment.
2025-03-11 20:39:23,417 - OrtInferSession - INFO: Second, install onnxruntime-directml by `pip install onnxruntime-directml`
INFO:OrtInferSession:Second, install onnxruntime-directml by `pip install onnxruntime-directml`
2025-03-11 20:39:23,417 - OrtInferSession - INFO: Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
INFO:OrtInferSession:Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
2025-03-11 20:39:26,510 - OrtInferSession - WARNING: DmlExecutionProvider is not in available providers (['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']). Use TensorrtExecutionProvider inference by default.
WARNING:OrtInferSession:DmlExecutionProvider is not in available providers (['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']). Use TensorrtExecutionProvider inference by default.
2025-03-11 20:39:26,525 - OrtInferSession - INFO: If you want to use DirectML acceleration, you must do:
INFO:OrtInferSession:If you want to use DirectML acceleration, you must do:
2025-03-11 20:39:26,525 - OrtInferSession - INFO: First, uninstall all onnxruntime pakcages in current environment.
INFO:OrtInferSession:First, uninstall all onnxruntime pakcages in current environment.
2025-03-11 20:39:26,527 - OrtInferSession - INFO: Second, install onnxruntime-directml by `pip install onnxruntime-directml`
INFO:OrtInferSession:Second, install onnxruntime-directml by `pip install onnxruntime-directml`
2025-03-11 20:39:26,527 - OrtInferSession - INFO: Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
INFO:OrtInferSession:Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
2025-03-11 20:39:26,564 - OrtInferSession - WARNING: DmlExecutionProvider is not in available providers (['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']). Use TensorrtExecutionProvider inference by default.
WARNING:OrtInferSession:DmlExecutionProvider is not in available providers (['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']). Use TensorrtExecutionProvider inference by default.
2025-03-11 20:39:26,566 - OrtInferSession - INFO: If you want to use DirectML acceleration, you must do:
INFO:OrtInferSession:If you want to use DirectML acceleration, you must do:
2025-03-11 20:39:26,566 - OrtInferSession - INFO: First, uninstall all onnxruntime pakcages in current environment.
INFO:OrtInferSession:First, uninstall all onnxruntime pakcages in current environment.
2025-03-11 20:39:26,566 - OrtInferSession - INFO: Second, install onnxruntime-directml by `pip install onnxruntime-directml`
INFO:OrtInferSession:Second, install onnxruntime-directml by `pip install onnxruntime-directml`
2025-03-11 20:39:26,568 - OrtInferSession - INFO: Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
INFO:OrtInferSession:Third, ensure DmlExecutionProvider is in available providers list. e.g. ['DmlExecutionProvider', 'CPUExecutionProvider']
INFO:docling.utils.accelerator_utils:Accelerator device: 'cuda:0'...`

### Docling version

... docling version - 2.26.0

### Python version

...python - 3.12.7