fix(core): implement sleep delay in FakeMessagesListChatModel _generate

[siddarthreddygsr]

implement sleep delay in FakeMessagesListChatModel._generate so the sleep parameter is respected, matching the documented behavior. This adds artificial latency between responses for testing purposes.

Issue: closes #31974 following docs

Dependencies: none

Twitter handle: @siddarthreddyg2

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
langchain	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Jul 18, 2025 6:12pm

[codspeed-hq]

CodSpeed WallTime Performance Report

Merging #32014 will not alter performance

_{Comparing siddarthreddygsr:master (8711c52) with master (50a12a7)}

⚠️ 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

✅ 13 untouched benchmarks

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #32014 will not alter performance

_{Comparing siddarthreddygsr:master (8711c52) with master (50a12a7)}

Summary

✅ 14 untouched benchmarks

[mdrxy]

LGTM, @eyurtsev for final approval

[qwikgig-krishna]

This looks like a good fix. Have you considered adding a test case to verify that the sleep parameter is respected? This would help prevent regressions in the future.

[mdrxy]

+1 on @qwikgig-krishna, some elementary test would be good here

[siddarthreddygsr]

Added the test case @mdrxy