ai summary added

[parinaB]

Description

The /explanation/ endpoint was returning a hardcoded template summary like
"A short Python snippet (3 lines) that performs a focused task." — the engine
never read the actual code. This PR routes the summary field through the
existing ai_provider.py LLM abstraction when LLM_ENABLED=true, with
graceful fallback to the rule-based template when LLM is disabled or
unavailable.

Before:

"summary": "A short Python snippet (3 lines) that performs a focused task. Good starting point for learners."

After:

"summary": "The function `calculate_tax` computes a 30% tax on the given income and returns the result."

Related Issue

Fixes #1065

Type of change

• Bug fix
• New feature / enhancement
• Documentation update
• Test addition
• Refactor

Checklist

• I have read CONTRIBUTING.md
• My branch is up to date with main
• I have run pytest -v and all tests pass
• I have not introduced duplicate issues or features
• My PR title follows the format: feat/fix/docs/test: short description
• [] I have added tests for new features (Level 2 and 3 issues)
• No hardcoded secrets or API keys in my code
• This PR is linked to a GSSoC 2026 issue

Test evidence

Files changed:

• backend/app/main.py — added load_dotenv() before all imports
• backend/app/services/code_assistant.py — run_explanation() calls call_llm() for summary when LLM is enabled

Note: Production deployments default to rule-based — LLM is opt-in via .env. No secrets committed.

Frontend correctly displays the AI-generated summary when the backend has LLM_ENABLED=true. No frontend changes needed. The deployed demo at qyverixai.onrender.com shows rule-based output by default as Render env vars are not configured — contributors running locally with a valid .env will see AI summaries.

[imDarshanGK]

@parinaB resolve conflicts and Please add a video demo of the feature.