Unnecessary permissions on iOS when using on-device recognition

[msschwartz]

If user is using the on-device recognition, we do not need to prompt the user for speech recognition permission.

Speech data from this app will be sent to Apple to process your requests. This will also help Apple improve its speech recognition technology.

This can be a show stopper on apps with sensitive data that users prefer to keep private.

A quick proof-of-concept PR has been created to show how this can be improved to separate the microphone permissions from speech recognizer permissions. https://github.com/jamsch/expo-speech-recognition/pull/55

[jamsch]

Thank you @msschwartz! Yep, I overlooked this and we definitely need more granular permissions here in cases like this, as well as for troubleshooting permissions issues. I'll be able to review your PR later today but my initial impressions are:

• Implement requestMicrophonePermissionsAsync() to request RECORD_AUDIO permissions for Android and (maybe) suffix the method name for requestSpeechRecognizerPermissionsAsync() with IOS (if no other platform has an equivalent permission).
• Probably still keep the requestPermissionsAsync API around without the deprecation marker just so that it's the easy path for most (as Android only needs one permission).