nvme_test: cc.enable() delay capability during servicing for the NvmeFaultController

[gurasinghMS]

This PR will add a fault functionality while changing the cc enable bit of the nvme fault controller

[Copilot]

Pull Request Overview

This PR adds fault injection capabilities for PCI controller management operations in the NvMe test framework, specifically enabling delay faults during the cc.enable() process. The changes allow testing of NvMe controller behavior when delays are introduced during controller enablement operations.

Key changes:

• Introduces a new PciFaultConfig struct with support for delay behavior during cc.enable() operations
• Updates all test instantiations to include the new PCI fault configuration
• Implements fault processing logic in the controller's cc.enable() handler

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
vm/devices/storage/nvme_resources/src/fault.rs	Adds new PciFaultConfig and PciFaultBehavior types for PCI fault injection
vm/devices/storage/nvme_test/src/pci.rs	Implements fault handling in the cc.enable() logic and stores fault configuration
vm/devices/storage/nvme_test/src/tests/controller_tests.rs	Updates test instantiations to include PCI fault configuration
vm/devices/storage/nvme_test/src/tests/shadow_doorbell_tests.rs	Updates test instantiations to include PCI fault configuration
vm/devices/storage/disk_nvme/nvme_driver/src/tests.rs	Updates test instantiations to include PCI fault configuration
vmm_tests/vmm_tests/tests/tests/multiarch/openhcl_servicing.rs	Updates test instantiations and improves error message clarity

Comments suppressed due to low confidence (1)

vm/devices/storage/nvme_resources/src/fault.rs:9

• The Duration import is used by the new PciFaultBehavior::Delay variant but there's no documentation explaining the expected duration ranges or behavior. Consider adding documentation to clarify appropriate delay values for testing scenarios.

use std::time::Duration;

[gurasinghMS]

@mattkur I am adding another fault type enum for the PCI style faults as we discussed earlier. A little unsure of the naming here because technically speaking cc.en() is still an NVMe concept. I am a little unsure if calling this a PCI fault is the right thing to do.
TODO: Fault configuration should also have a builder-style implementation. Will make that a follow up PR to this one.

[mattkur]

Thanks! Re: naming: I'm open to other names, but this is the pci interface of the NVMe controller. Since it is scoped to the NVMe concepts by crate and module, this makes sense to me.