Krutrim-cloud

[Text-to-Text Models]
- Added support of **Gemma-2-27B-IT** & **Reflection-Llama-3.1-70B** for ChatCompletion API

[Text-to-Image Model]
- Updated input validation for input height and width of stable diffusion model parameters [min:256, max:2048]

Audio Models
* New Features:
- **Extended Audio Processing: Now supports processing of longer audio files more than 6 hours.**
- **Video Input Support: Added functionality to handle video files by automatically converting video to audio for processing.**
- **Audio Validation**: Introduced a validation mechanism to ensure audio integrity and format compliance.
- **Audio Manipulation**: Implemented support for cropping and concatenating audio files using PyDub, with export capabilities to local storage.

Image Models
* Examples Added:
- Support for the OpenCV library.
- Reading and saving images in various color spaces: RGB, RGBA, CMYK.
- Image resizing with multiple algorithms.
- Image manipulation functionalities including:
* Rotation
* Flipping
* Grayscale conversion
* Brightness and contrast adjustment

Text Models
* New Features:
- Added a Prompt Registry with example usage for enhanced text model prompting.

Initial release of SDK supporting Model Inference APIs from Krutrim Cloud

**Supported Models:**

Text to Text:
krutrim/Krutrim-spectre-v2
meta/Llama-3-8B-Instruct
mistralai/Mistral-7B-v0.2

Audio to Text:
openai/whisper-large-v3

Text to Image:
stabilityai/stable-diffusion-xl-base-1.0

Image to Text:
HuggingFaceM4/idefics2-8b-base

**Full Changelog**: https://github.com/ola-krutrim/krutrim-cloud-python/commits/v0.1.0