feat(retrieval): improve search query preprocessing (#162)

[prasiddhi-105]

Description

• Introduced search query preprocessing to normalize raw user input before hitting the retrieval pipeline.
• Normalized irregular whitespace (newlines, tabs, redundant spaces) and stripped unnecessary surrounding punctuation while preserving essential code symbols.
• Maintained compatibility with existing semantic and hybrid retrieval systems.

---

Related Issue

Fixes #162

---

Type of Change

• Bug fix
• New feature
• Enhancement
• Documentation update
• Refactoring
• Performance improvement
• Tests
• Other (please specify)

---

Changes Made

• Created preprocess_query() helper function in retriever.py.
• Integrated automatic query preprocessing inside retrieve() entry point.
• Added unit tests in tests/test_retriever.py covering whitespace normalization and symbol preservation.

---

Testing

• Tested locally
• Existing tests pass
• Added new tests (if applicable)

Additional Notes:
Ran python -m pytest tests/test_retriever.py locally; all tests passed cleanly.

---

Screenshots (if applicable)

N/A - Backend preprocessing logic.

---

Checklist

• My code follows the project's coding style.
• I have linked the related issue.
• I have tested my changes.
• I have updated documentation where necessary.
• My changes do not introduce new warnings or errors.
• The project builds successfully.
• This PR only addresses the linked issue.

[Aharshi3614]

@prasiddhi-105 The build is failing here. Can you please fix this