[FEATURE] Confidence score visualization

[Rudra-clrscr]

Problem

The system currently returns only a classification label (Spam/Smishing/Offensive/Ham) without indicating how certain the model is, making it hard for users to judge borderline cases.

Proposed Solution

Display a confidence score/probability percentage alongside each classification result, with a simple visual indicator (e.g., a progress bar or color-coded badge: high/medium/low confidence).

Acceptance Criteria

• Classification result includes a confidence percentage
• Visual indicator (bar/badge) reflecting confidence level
• Low-confidence results are visually distinguished (e.g., "Uncertain - please review")
• Confidence score is included in API responses as well as the UI

[github-actions]

👋 Welcome back! Thanks for your continued contribution. Your issue/PR is received and waiting for maintainer review.

[github-actions]

The request proposes adding a confidence score to classification results (e.g., Spam, Smishing, Offensive, Ham) to help users evaluate borderline cases. The solution involves displaying a percentage and a visual indicator (such as a progress bar or color-coded badge) to reflect confidence levels, emphasizing low-confidence results for user review. This feature would be integrated into both the UI and API responses.

[Userunknown84]

Good approach but this already assigned to someone else

[Rudra-clrscr]

@Userunknown84 OK