feat(anthropic,model-profiles): add effort support

[mdrxy]

model-profiles

• Add new reasoning_effort_control to capture whether the model supports controlling reasoning effort. This is what I interpret the new effort param to mean, but this is debatable and I'm open to backtracking

  "Use high effort (the default) when you need Claude's best work—complex reasoning, nuanced analysis, difficult coding problems, or any task where quality is the top priority"

anthropic

• Add named effort param
• If output_config is provided as a kwarg, that works too
• Automatically append the required beta for effort if it is detected in either
• Add tests

[codspeed-hq]

CodSpeed Performance Report

Merging #34116 will improve performances by 27.51%

_{Comparing mdrxy/effort-anthropic (43a2253) with mdrxy/anthropic-auto-append-betas (1ab885d)}

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

Summary

⚡ 7 improvements
✅ 9 untouched
⏩ 18 skipped¹

Benchmarks breakdown

	Mode	Benchmark	BASE	HEAD	Change
⚡	WallTime	test_async_callbacks_in_sync	23.5 ms	18.4 ms	+27.51%
⚡	WallTime	test_import_time[CallbackManager]	456 ms	410 ms	+11.23%
⚡	WallTime	test_import_time[HumanMessage]	268.3 ms	242.7 ms	+10.56%
⚡	WallTime	test_import_time[LangChainTracer]	433.8 ms	391.3 ms	+10.86%
⚡	WallTime	test_import_time[PydanticOutputParser]	529 ms	470.9 ms	+12.34%
⚡	WallTime	test_import_time[Runnable]	489 ms	443.9 ms	+10.15%
⚡	WallTime	test_import_time[tool]	512.8 ms	458.3 ms	+11.88%

Footnotes

1. 18 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩