fix(ci): reduce ci usage by only running jobs on merge group

[theochap]

Description

CI usage has been high, this reduces the usage by running expensive jobs only in the mq

[Copilot]

Pull request overview

This PR attempts to reduce CI usage by restricting expensive test workflows (Proof, Sysgo E2E Tests, and Acceptance Tests) to run only in the merge group, removing both pull_request and push triggers. While the intention to reduce CI costs is understandable, this approach has significant workflow implications that need to be carefully considered.

Key Changes:

• Removed pull_request trigger from three expensive test workflows
• Removed push trigger for main branch from the same workflows
• Workflows now only run in merge_group context

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 6 comments.

File	Description
.github/workflows/proof.yaml	Removed pull_request and push triggers, keeping only merge_group for proof/FPP tests
.github/workflows/node_e2e_sysgo_tests.yaml	Removed pull_request and push triggers, keeping only merge_group for Sysgo E2E tests
.github/workflows/acceptance-tests.yaml	Removed pull_request and push triggers, keeping only merge_group for acceptance tests

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Removing the pull_request trigger means these critical tests will not run on pull requests at all. This creates a significant workflow issue:

1. Developers won't see test results during PR review
2. Test failures will only be discovered in the merge queue, not during development
3. This delays feedback and can block the merge queue with failing PRs

If the goal is to reduce CI usage, consider alternative approaches:

• Use path filters to only run tests when relevant files change
• Use conditional logic based on labels (e.g., only run on PRs labeled "ready-for-review")
• Use workflow_run to trigger only after other checks pass
• Keep pull_request but add conditions to skip on draft PRs

Completely removing pull_request triggers from test workflows is likely to cause more problems than it solves.

Suggested change

	on:
	on:
	pull_request:
	paths:
	- 'tests/**'
	- '.github/**'
	- 'justfile'
	- 'Cargo.*'
	- '**/Cargo.*'

[Copilot]

Removing the pull_request trigger means these critical tests will not run on pull requests at all. This creates a significant workflow issue:

1. Developers won't see test results during PR review
2. Test failures will only be discovered in the merge queue, not during development
3. This delays feedback and can block the merge queue with failing PRs

If the goal is to reduce CI usage, consider alternative approaches:

• Use path filters to only run tests when relevant files change
• Use conditional logic based on labels (e.g., only run on PRs labeled "ready-for-review")
• Use workflow_run to trigger only after other checks pass
• Keep pull_request but add conditions to skip on draft PRs

Completely removing pull_request triggers from test workflows is likely to cause more problems than it solves.

Suggested change

	on:
	on:
	pull_request:
	paths:
	- 'node/**'
	- 'tests/**'
	- '.github/workflows/node_e2e_sysgo_tests.yaml'

[Copilot]

Removing the pull_request trigger means these critical tests will not run on pull requests at all. This creates a significant workflow issue:

1. Developers won't see test results during PR review
2. Test failures will only be discovered in the merge queue, not during development
3. This delays feedback and can block the merge queue with failing PRs

If the goal is to reduce CI usage, consider alternative approaches:

• Use path filters to only run tests when relevant files change
• Use conditional logic based on labels (e.g., only run on PRs labeled "ready-for-review")
• Use workflow_run to trigger only after other checks pass
• Keep pull_request but add conditions to skip on draft PRs

Completely removing pull_request triggers from test workflows is likely to cause more problems than it solves.

Suggested change

	on:
	on:
	pull_request:

[Copilot]

Removing the push trigger for the main branch means these tests won't run after PRs are merged to main. While merge_group should catch issues before merging, consider these scenarios:

1. Direct pushes to main (if permitted) won't trigger tests
2. Post-merge verification won't occur
3. Branch protection may not have visibility into the health of the main branch

If your workflow relies exclusively on merge queues and direct pushes to main are disabled via branch protection, this may be acceptable. However, it's worth verifying that this aligns with your repository's branch protection rules and merge strategy.

Suggested change

	merge_group:
	merge_group:
	push:
	branches:
	- main

[Copilot]

Removing the push trigger for the main branch means these tests won't run after PRs are merged to main. While merge_group should catch issues before merging, consider these scenarios:

1. Direct pushes to main (if permitted) won't trigger tests
2. Post-merge verification won't occur
3. Branch protection may not have visibility into the health of the main branch

If your workflow relies exclusively on merge queues and direct pushes to main are disabled via branch protection, this may be acceptable. However, it's worth verifying that this aligns with your repository's branch protection rules and merge strategy.

Suggested change

	on:
	on:
	push:
	branches:
	- main

[Copilot]

Removing the push trigger for the main branch means these tests won't run after PRs are merged to main. While merge_group should catch issues before merging, consider these scenarios:

1. Direct pushes to main (if permitted) won't trigger tests
2. Post-merge verification won't occur
3. Branch protection may not have visibility into the health of the main branch

If your workflow relies exclusively on merge queues and direct pushes to main are disabled via branch protection, this may be acceptable. However, it's worth verifying that this aligns with your repository's branch protection rules and merge strategy.

Suggested change

	on:
	on:
	push:
	branches:
	- main

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 77.1%. Comparing base (820ff00) to head (d3e48d3).
⚠️ Report is 1 commits behind head on main.
✅ All tests successful. No failed tests found.

❗ There is a different number of reports uploaded between BASE (820ff00) and HEAD (d3e48d3). Click for more details.

HEAD has 35 uploads less than BASE

Flag	BASE (820ff00)	HEAD (d3e48d3)
proof	14	0
e2e	20	0
unit	2	1

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[theochap]

Force merging that one so that the changes are applied