docs: clarify least-squares intuition in linear regression lesson

[dagimtessema]

Description

This PR improves the explanation of the "Least-Squares" method in the Linear Regression README.

Motivation

While going through the lesson as a student, I found the existing explanation of why we square the residuals to be a bit brief. I have rephrased it to more clearly explain:

1. Why we use magnitude instead of direction (avoiding negative values).
2. The goal of minimizing the final sum to find the "best fit" line.

I believe this version is more accessible to beginners who are encountering the concept for the first time.

Type of change

• Documentation update (non-breaking change)

[Copilot]

Pull request overview

This PR improves the documentation for the Least-Squares Regression concept in the Linear Regression lesson by expanding and clarifying the explanation of why residuals are squared. The revised content provides a more structured, beginner-friendly explanation with numbered points and clearer reasoning.

Changes:

• Rephrased the introduction to Least-Squares Regression to explicitly mention residuals and the goal of minimizing total error
• Added a numbered list explaining two main reasons for squaring distances: avoiding negative cancellation and penalizing outliers
• Improved the closing explanation to emphasize finding the line with the smallest sum

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The numbered list items have incorrect spacing. There should be a space between the number and the period (e.g., "1." not "1.") and the text. The current formatting "1.**Magnitude" should be "1. **Magnitude". The same applies to item 2.

Suggested change

	1.**Magnitude over Direction:** We want to treat an error of -5 the same as an error of +5. Squaring turns all values positive.
	2.**Penalizing Outliers:** Squaring gives more weight to larger errors, forcing the line to stay closer to points that are far away.
	1. **Magnitude over Direction:** We want to treat an error of -5 the same as an error of +5. Squaring turns all values positive.
	2. **Penalizing Outliers:** Squaring gives more weight to larger errors, forcing the line to stay closer to points that are far away.

[dagimtessema]

@copilot open a new pull request to apply changes based on the comments in this thread