feat: Add MLX delegate for whisper

[IgorSwat]

Description

This PR registers MLX whisper models and sets them up as default for a physical iOS device.
On average, MLX delegates run 50% faster than current CoreML delegates for this model, while keeping the same model size.

Introduces a breaking change?

• Yes
• No

Type of change

• Bug fix (change which fixes an issue)
• New feature (change which adds functionality)
• Documentation update (improves or adds clarity to existing documentation)
• Other (chores, tests, code style improvements etc.)

Tested on

• iOS
• Android

Testing instructions

Run speech-to-text demo app to check out if MLX delegates are downloaded and run correctly.

Screenshots

Related issues

Checklist

• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have updated the documentation accordingly
• My changes generate no new warnings

Additional notes

[barhanc]

@IgorSwat could you also add the dynamic input companion methods to Whisper exports as part of this, since I see the new models were already added to the v0.10.0 tag.

[IgorSwat]

@IgorSwat could you also add the dynamic input companion methods to Whisper exports as part of this, since I see the new models were already added to the v0.10.0 tag.

Yeah, good catch. I completely forgot about it 😅