v0.30.0

Major Changes

• feat(cuda): add CUDA 13.0.0 jobs to the compile workflow by @modelpath-dev in #1439
• feat(cuda): probe cuda13 native libraries before cuda12 by @modelpath-dev in #1440
• feat(cuda): add LLamaSharp.Backend.Cuda13 packages by @modelpath-dev in #1441
• Add Microsoft Agent Framework (Microsoft.Agents.AI) with RAG Example by @zsogitbe in #1423
• Expose llama_model_params.devices as IModelParams.Devices by @ipevo-brooks in #1447

Bug Fixes

• Detect Vulkan via the loader before falling back to vulkaninfo by @ipevo-brooks in #1446
• Remove SemanticKernel and KernelMemory leftovers by @m0nsky in #1426
• Disable SSE4.2 in noavx native builds by @modelpath-dev in #1438
• Fix RawRepresentationFactory settings being overwritten in AsChatClient by @Laurianti in #1442
• Fixed CI by @martindevans in #1458

Other Changes

• Read the release version from LLamaSharp.csproj by @m0nsky in #1427
• Merge the minor and patch release workflows into one by @m0nsky in #1430
• Push the API references with GITHUB_TOKEN, falling back to a pull request by @Laurianti in #1453
• Fix the API references workflow for XMLDoc2Markdown 6 by @Laurianti in #1443
• Build the API references from net10.0 by @Laurianti in #1462
• Fixed release-trigger.yml by @martindevans in #1464
• Big package update by @martindevans in #1457
• Cleanup Unused Things by @martindevans in #1459

New Contributors

• @modelpath-dev made their first contribution in #1438
• @Laurianti made their first contribution in #1442
• @ipevo-brooks made their first contribution in #1446

Full Changelog: v0.29.0...v0.30.0