Fix/stt output

[Khurdhula-Harshavardhan]

This pull request updates the package version and modifies the structure of the language detection output in the speech-to-text response. The most important changes are listed below.

Version update:

• Updated the package version to 0.4.1 in both setup.py and jigsawstack/version.py for release tracking and distribution. [1] [2]

Speech-to-text API changes:

• Changed the language_detected field in the SpeechToTextResponse class from an optional string to an optional dictionary, allowing for richer language detection metadata in the transcription response.
• Removed the confidence field from SpeechToTextResponse, simplifying the response and removing language detection confidence reporting.

NOTE: this is a breaking change for STT, for users using jigsawstack-python under and equal 0.4.0, when pointed, idea is to suggest the user to upgrade to 0.4.1 latest