Release/0.3.7

[JannikSt]

No description provided.

[Copilot]

Pull Request Overview

This PR prepares the 0.3.7 release by extending timeouts, exposing a new Docker network‐mode flag, adding execution‐time metrics, and introducing grace‐period logic for discovery.

• Increased timeouts and expirations (heartbeat client 5→20s, Redis 60→90s).
• Added a disable_host_network_mode option to DockerService/Manager and exposed it in the CLI.
• Introduced metrics for status‐update execution time and a 5-minute grace period before ejecting nodes.

Reviewed Changes

Copilot reviewed 12 out of 13 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
crates/worker/src/operations/heartbeat/service.rs	Increased HTTP client timeout to 20s
crates/worker/src/docker/service.rs	Added disable_host_network_mode parameter to DockerService::new
crates/worker/src/docker/docker_manager.rs	Integrated disable_host_network_mode into DockerManager
crates/worker/src/cli/command.rs	Exposed --disable-host-network-mode flag
crates/validator/src/validators/hardware.rs	Switched to sequential node validation to avoid nonce conflicts
crates/shared/src/security/auth_signature_middleware.rs	Introduced REQUEST_EXPIRY_SECS constant
crates/orchestrator/src/store/domains/heartbeat_store.rs	Extended Redis key expiration from 60s to 90s
crates/orchestrator/src/status_update/mod.rs	Injected MetricsContext, recording per-node execution time
crates/orchestrator/src/metrics/mod.rs	Added status_update_execution_time histogram and recorder
crates/orchestrator/src/main.rs	Passed metrics context to status updater
crates/orchestrator/src/discovery/monitor.rs	Added a 5-minute grace period before marking nodes inactive
Cargo.toml	Bumped workspace version to 0.3.7

Comments suppressed due to low confidence (1)

crates/orchestrator/src/status_update/mod.rs:323

• The new metrics recording for status-update execution time isn’t covered by tests. Consider adding unit or integration tests to assert that record_status_update_execution_time behaves as expected.

            self.metrics.record_status_update_execution_time(