Budget-tree consumer-0 + Track A: instantiate the real srv1 fleet BudgetNode tree from measured numbers + DERIVE the .dag-floor spawn-width from it (construction, not a Bool check)

[briansrls]

Auto-opened by session-dashboard for session calm-badger-518.
Pushing to session/calm-badger-518 advances this PR.

Worker attestation

Before flipping this PR to ready for review, confirm each item:

• Title describes the change (not the session id or branch).
• PR body summarises what and why (replace the TODO below).
• Tests run: name the command (e.g. npm test, cargo test) and the result.
• If this closes a work item, the body contains a Closes #N directive.
• No commits on this branch are surprises (no fork/cherry-pick I did not make).
• No secrets / credentials / large binaries staged.

Summary

TODO: replace this paragraph with one or two sentences naming the change and its motivation. Reviewers read this first.

Test plan

• TODO: list the commands that ran (or "no tests changed; relied on CI") and the outcome.

[gunbai-bot]

Closing as a duplicate of #5590 (the consumer-0 + Track A work-item auto-spawned two children; both produced a full PR). Consolidating on #5590 because it is the better foundation:

• Contract-compliant tree. Budget-tree consumer-0 + Track A: instantiate the real srv1 fleet BudgetNode tree from measured numbers + DERIVE the .dag-floor spawn-width from it (construction, not a Bool check) #5590 keeps the L1 run-pool overhead-only (byte_size_saturating_sub(host, overhead)), per the ratified "tree carries no margin; the single 0.8 lives consumer-side" contract. This PR instead bakes a tree-side 0.8 into the run-pool via memory_effective_budget, which would double-margin once Track B's placement (R from L1) applies its own consumer-side 0.8.
• Reusable descent primitive. Budget-tree consumer-0 + Track A: instantiate the real srv1 fleet BudgetNode tree from measured numbers + DERIVE the .dag-floor spawn-width from it (construction, not a Bool check) #5590 adds budget_child_named (a closed-sum BudgetChild descent) that both Track B lanes (R from L1, N from L2) read their allocation through; this PR uses local accessors that don't generalize.

No work is lost — #5590 carries the same measured-carrier instantiation and the same DERIVE-not-check floor spawn-width. Thanks for the reviews here; they applied to the shared design and informed #5590.

— sent from quick-ant-298