Fix/chat fallback code correction

[Void-Anvesha]

Description

Fixes the follow-up chat fallback response so it provides clear and actionable code corrections when the LLM is unavailable.

The chat assistant now:

• Detects Python syntax errors.
• Identifies the affected line.
• Explains what needs to be changed.
• Displays corrected code.

Example input:

print(hello);;

The response explains the extra semicolon, advises adding quotes for text, and provides:

print("hello")

The frontend now prefers the local backend (http://localhost:8000) when opened locally, so local changes are reflected immediately.

Related Issue

Fixes #

Type of change

• Bug fix
• New feature / enhancement
• Test addition

Checklist

• I have read CONTRIBUTING.md
• My branch is up to date with main
• I have run relevant validation
• I have added tests for the new behavior

Test evidence

Verified the local /chat/ endpoint with:

print(hello);;

The response returns the issue, the required changes, and corrected code.

[Void-Anvesha]

#1674
https://drive.google.com/file/d/1HYvMyHerMN3jFuZZHxRrRMmG5F16rV8V/view?usp=sharing
This is the drive link for the feature that I've added on.
Review this & accordingly merge the changes

[imDarshanGK]

@Void-Anvesha issue number?

[Void-Anvesha]

#1659

[imDarshanGK]

@Void-Anvesha resolve conflicts