Reasoning settings now resolve portably across AI SDK providers — effort levels and enable/disable flags map to the AI SDK's native reasoning setting (including Ollama), replacing the per-provider thinking overrides, and an explicit request to disable reasoning now takes priority `sdk.error` telemetry from agent runs is attributed to the model actually in use, and undefined values are stripped from the event properties Refreshed the model catalog from models.dev, and surfaced `meta/muse-spark-1.2-contributor` for the Cline provider Full Changelog: https://github.com/cline/cline/compare/sdk/sdk/v0.0.70...sdk/sdk/v0.0.71