fix: reconcile hosted compute capacity updates

[alongubkin]

Summary

• route hosted compute changes through the normal Platform update lifecycle
• allow registered compute controllers to reconcile capacity and scale-policy changes without reopening setup-owned cloud boundaries
• preserve omitted setup-owned resources and reject changes to machine profiles, placement, networking, and other frozen configuration

Verification

• frozen-resource compatibility tests
• deployment update and retry tests
• generated CLI hosted-lifecycle tests

[greptile-apps]

Greptile Summary

The PR routes hosted compute-capacity changes through the deployment update lifecycle while preserving setup-owned frozen resources across consecutive releases.

• Extends hosted compute updates to active and retryable deployment states.
• Allows narrowly scoped ComputeCluster capacity and scale-policy changes while retaining frozen structural boundaries.
• Carries omitted frozen resources into the effective update target and durable prepared-stack baseline.
• Adds lifecycle, compatibility, retry, and ownership-preservation coverage.

Confidence Score: 5/5

The PR appears safe to merge because no blocking failure remains in the eligible follow-up scope.

No blocking failure remains.

Important Files Changed

Filename	Overview
crates/alien-deploy-cli/src/commands/up.rs	Expands hosted compute-update routing to active and retryable lifecycle states and adds predicate coverage.
crates/alien-deployment/src/updating.rs	Preserves omitted frozen resources in update targets, persists the effective baseline, and includes frozen resources in managed execution.
crates/alien-deployment/tests/test_platform.rs	Verifies omitted setup-owned frozen resources survive consecutive release updates.
crates/alien-preflights/src/compatibility/frozen_resources_unchanged.rs	Permits only ComputeCluster capacity and scale-policy changes while continuing to reject structural frozen-resource modifications.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart LR
    A[Hosted compute settings change] --> B[Deployment update lifecycle]
    B --> C[Compatibility preflights]
    C --> D{Change type}
    D -->|Capacity or scale policy| E[Managed frozen reconciliation]
    D -->|Machine profile, placement, networking, or identity| F[Reject and require setup]
    E --> G[Merge omitted frozen resources]
    G --> H[Execute live and eligible frozen resources]
    H --> I[Promote effective prepared-stack baseline]

Loading

_{Reviews (4): Last reviewed commit: "fix: retain frozen resources across upda..." | Re-trigger Greptile}