feat: add Valkey support (storage db + vector db)

[atao2004]

Summary

Adds full Valkey support to Agno: a ValkeyDb storage adapter, a ValkeyDB vector database, cookbooks, and unit/integration tests. Both clients set a client_name so connections are identifiable via CLIENT LIST on the server.

Closes #8117, closes #7795

Storage adapter (agno/db/valkey)

• ValkeyDb implementing the full BaseDb interface: sessions, user memories, metrics, evals, knowledge, cultural knowledge, traces, and spans.
• Built on valkey-glide-sync (GlideClient / GlideClusterClient), with support for host/port, username/password auth, TLS, logical database selection, or a pre-configured client instance.
• Secondary index sets per record type (e.g. user_id, agent_id, session_type, trace_id) for filtered lookups without full scans.
• Bulk operations (upsert_sessions, upsert_memories, delete_sessions, create_spans, delete_eval_runs) batched through GLIDE pipelines, with per-record write verification.
• Optional key TTL via expire.
• Connections are named agno_db_client (CLIENT SETNAME).

Vector database (agno/vectordb/valkey)

• ValkeyDB (exported alias ValkeyVectorDb) implementing the VectorDb interface using the valkey-search module (FT.* commands).
• Vector search (KNN over HNSW or FLAT index; cosine, L2, and inner-product distance) and keyword search, plus reranker support.
• Sync and async variants for all interface methods.
• Insert, upsert (replace by content hash), delete by id/name/metadata/content_id, metadata updates, and existence checks (id, name, content hash).
• Connections are named agno_vectordb_client.

Cookbooks

• cookbook/06_storage/valkey/ — agent, team, and workflow storage examples with README.
• cookbook/05_agent_os/dbs/valkey_db.py — AgentOS backed by Valkey.
• cookbook/07_knowledge/05_integrations/vector_dbs/05_valkey.py — knowledge base with Valkey as the vector store.

Tests

• Unit tests for the storage adapter and vector database (run without a server via a stubbed GLIDE client).
• Integration tests for bulk operations, trace/span stats, and vector database operations against a live Valkey instance.

Dependencies

• New valkey extra: valkey-glide-sync>=2.1.0, included in the storage and vectordbs extras.

Run Valkey locally with:

docker run --name my-valkey -p 6379:6379 -d valkey/valkey-bundle

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)

Duplicate and AI-Generated PR Check

• I have searched existing open pull requests and confirmed that no other PR already addresses this issue
• If a similar PR exists, I have explained below why this PR is a better approach
• Check if this PR was entirely AI-generated (by Copilot, Claude Code, Cursor, etc.)

---

Additional Notes

• Hybrid search is not supported by the Valkey vector database; supported search types are vector and keyword.
• The valkey/valkey-bundle Docker image is required for vector search (it ships the valkey-search module); plain valkey/valkey is sufficient for the storage adapter.
• Learning methods are not implemented, matching the Redis adapter.

[kausmeows]

traces filtering from FE filter bar is not working-

[harshsinha03]

Hi @atao2004, thanks for this addition! We'll get this merged soon.

Would you also be open to adding the docs for this here: https://github.com/agno-agi/docs?

[atao2004]

 Thanks for the review @kausmeows all feedback addressed in commit 3c659af:

[atao2004]

Hi @harshsinha03 added docs here agno-agi/docs#700 :)

[harshsinha03]

LGTM, thanks @atao2004 and @edlng for the great work here, really appreciate the contribution! 🙌