[fix] Apply reranker to Milvus search results (fixes #5190)

[aayush598]

Summary

This PR updates the Milvus.search method to apply a reranker to the search results, ensuring that results are reordered based on relevance after initial vector search.

Key changes:

• Added a check for self.reranker and non-empty search results.
• Applied self.reranker.rerank(query=query, documents=search_results) to reorder results.
• Enforced limit on the final results after reranking.
• Updated libs/agno/tests/unit/vectordb/test_milvusdb.py with a unit test validating the reranker integration.

This resolves the reported issue: [#5190]

Type of change

• Bug fix
• New feature
• Breaking change
• Improvement
• Model update
• Other:

---

Checklist

• Code complies with style guidelines
• Ran format/validation scripts (./scripts/format.sh and ./scripts/validate.sh)
• Self-review completed
• Documentation updated (comments, docstrings)
• Examples and guides: Relevant cookbook examples have been included or updated (if applicable)
• Tested in clean environment
• Tests added/updated (if applicable)

---

Additional Notes

• The reranker is optional; if self.reranker is not defined, the search behaves as before.
• The search limit is enforced after reranking to ensure returned results respect the limit parameter.
• Unit test uses a mocked reranker that reverses results to simulate reranking behavior.

[manuhortet]

Thanks for the contribution @aayush598, we really appreciate it! 💪