v0.4.0

Upgrades to support Transformers 5, ollama as an inference backend, and various bug fixes

What's Changed

• Update structured outputs spec for vLLM 0.12.0+ as old format has been deprecated and removed. by @kndtran in #99
• Support Ollama backend by @kndtran in #102
• fix: missing escape in regex for citation by @jakelorocco in #106
• Transformers 5 support by @frreiss in #113
• Add tests for query clarification by @kndtran in #111
• Change test_run_ollama to use a mock server by @kndtran in #115
• Debugging workflow issue by @frreiss in #124
• Fix issue with span begin/end computation by @yannisk2 in #126

New Contributors

• @jakelorocco made their first contribution in #106
• @yannisk2 made their first contribution in #126

Full Changelog: v0.3.6...v0.4.0