fix: warm child Toolset before flattening in Toolset.add()

[mahesh-sadupalli]

Fixes #12009

Summary

Toolset.add() silently dropped all tools when a lazy child Toolset (one that materialises its tools in warm_up()) was added to a parent that had not yet been warmed up.

The guard was:

if self._is_warmed_up and hasattr(tool, "warm_up"):
    tool.warm_up()

new_tools = [tool] if isinstance(tool, Tool) else list(tool)

Because self._is_warmed_up was False for a cold parent, the child was never warmed before list(tool) flattened it. The list was empty and the tools were silently lost — no exception raised.

Fix: warm every child Toolset unconditionally before flattening. The child's warm_up() is idempotent, so calling it on an already-warm child is safe. Plain Tool objects retain the existing behaviour: warmed immediately only when the parent is already warm.

if isinstance(tool, Toolset):
    # Always warm a child Toolset before flattening so lazily-loaded tools
    # are available regardless of whether the parent has been warmed yet.
    tool.warm_up()
elif self._is_warmed_up and hasattr(tool, "warm_up"):
    tool.warm_up()

Tests

Three regression tests added to TestToolsetWarmUp:

Test	Covers
test_add_lazy_toolset_to_cold_parent_retains_tools	Main bug: lazy child added to cold parent retains its tools
test_add_lazy_toolset_warm_up_is_idempotent	Already-warm child added to warm parent is not warmed again
test_add_plain_tool_to_cold_parent_does_not_warm_it	Regression guard: plain Tool added to cold parent is still not warmed eagerly

All 34 tests in test/tools/test_toolset.py pass.

[vercel]

@mahesh-sadupalli is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[github-actions]

Hi @mahesh-sadupalli, thanks a lot for your contribution! 🙏

We noticed that the Contributor License Agreement (CLA) check (license/cla) hasn't passed yet, so we've temporarily moved this PR to draft and paused the review assignment.

To get your PR reviewed, please sign the CLA via the link in the license/cla check below (or in the CLA bot comment). As soon as the check turns green, this PR will automatically be marked ready for review again and a reviewer will be re-assigned.

[sjrl]

Closing as duplicate of #12012