feat(eval): support automatic speech recognition metrics

Feature gap

winml eval does not register automatic-speech-recognition, so valid Wav2Vec2 CTC artifacts cannot be evaluated with a task metric through the CLI.

Reproduction

• Model: MahmoudAshraf/mms-300m-1130-forced-aligner
• Task: automatic-speech-recognition
• Artifacts: verified CPU FP32 and FP16 ONNX models with input_values [1,16000] -> logits [1,49,31]

Run either:

winml eval --schema --task automatic-speech-recognition

or:

winml eval -m <model.onnx> --model-id MahmoudAshraf/mms-300m-1130-forced-aligner --task automatic-speech-recognition --device cpu --ep cpu

Both are rejected because ASR is absent from the supported-task registry.

Requested support

Add an ASR evaluator/schema with raw-audio preprocessing and a defensible transcript/audio dataset contract. For CTC checkpoints, expose an appropriate metric such as WER; forced-alignment checkpoints may additionally need alignment/timestamp quality evaluation.

The referenced model card only says “forced alignment dataset” and does not identify a dataset repository, subset, split, or revision, so callers must be able to provide an explicit evaluation dataset rather than silently substituting one.

Start with the supported-task registry used by winml eval --schema and reproduce the rejection with the commands in the issue. Define the explicit audio/transcript dataset contract, add ASR schema and raw-audio evaluation with an appropriate CTC metric such as WER, and verify that both schema discovery and model evaluation accept the task without silently selecting a dataset.