build: add package for kernel-memory integration.

This reverts commit 091b8d58b3502a99b3bfbec9db457c92cc736beb, reversing
changes made to 9b2ca9cf8e.

This reverts commit b4b3ea9d99.

Prevent duplication of user prompts / chat history in ChatSession.

Just merge cuda and avx detection and change layout.

Context Size Autodetect

Delete Backend.Metal because is not needed anymore.
Do not include .metal in x86_64 binaries

Update LLama.Examples using Spectre.Console

Enhance framework compatibility

The way ChatSession.ChatAsync was using the provided methods
from a IHistoryTransform interface implementation created unexpected
duplication of the chat history messages. It also prevented loading
previous history into the current session.

Create different paths to different MacOS platforms.
Dynamically load the right library

Align with llama.cpp b1488

- Handle invalid choice in Examples.TryGetValue
- Add exit condition for choice "Exit"
- Display a rule before executing example
- Clear console after each iteration

📝 This change refactors the TestRunner.cs file. It handles invalid choices in Examples.TryGetValue, adds an exit condition for the choice "Exit", displays a rule before executing an example, and clears the console after each iteration.

Removed Old Targets From CI matrix

Removed some CI targets

Update README.md

Add the MacOS binary from the same run

GitHub Action Pipeline Improvements

feat(kernel-memory): avoid loading model twice.