fix(verify-module): per-rung timeout budgets VM_TMO_GRAPH / VM_TMO_COVERAGE / VM_TMO_JOURNEYS, defaults unchanged (TK-008)

[MendixMau]

Direct lane

What changed and why (one paragraph): verify-module.sh had hard-coded rung budgets, so a clean module whose conformance probes took about 2 s each reported a timeout. Three env overrides let an operator raise a budget instead of accepting a false timeout; the default numbers are unchanged, so nothing changes unless the variable is set. Probe batching, plan-claims caching and a Jaeger retry are left for a later PR.

Impact

Row	Answer
Process	none; an operator can raise a budget instead of accepting a timeout
Dependencies	verify-module.sh, CHANGELOG
Context cost	none: no routed or always-loaded file touched
Behaviour	n/a, script only
Runtime	unchanged by default (budgets are the same numbers)

Field evidence — defect TK-008 of a five-app lab field report (Mendix 11.15, mxcli v0.26.0, toolkit d16f578); reproduced against master 354669e by the scoped test.

• No client data (leak guard, private-citation check clean)
• Size cap: within 10 files / 400 lines
• Test tier: T1, scoped test
• CHANGELOG.md line appended, crediting the source

Fix: VM_TMO_GRAPH, VM_TMO_COVERAGE, VM_TMO_JOURNEYS, defaults unchanged. Test: bash -n plus tests/wave2/test-verify-ledger-hint.sh (4 pass); no timing test, the change is a default-preserving substitution.

🤖 Generated with Claude Code