[lanyue-llk]

Problem

PermissionStatus exposes onchange, but its generated prototype does not inherit from EventTarget. Instances therefore cannot use the standard addEventListener, removeEventListener, or dispatchEvent APIs.

Result

• Declare PermissionStatus as an EventTarget subclass.
• Install a listener store on each returned permission status object.
• Cover inheritance, instance isolation, listener removal, once, callback targets, and the non-constructible interface.

This PR only supplies the standard event interface. It does not claim that permission state changes are emitted automatically.

Validation

• cargo fmt --all -- --check
• Added permission_status_supports_inherited_event_target_listeners

This replaces the independent permissions portion of #1476. Its Web Fetch changes are superseded by #1501.

[github-actions]

Sequential Navigation Soak A/B

✅ HEAD completed the 200-navigation resilience and memory observation.

One browser process, one target, and one CDP session navigate CSDN → SegmentFault → Huaban → example.com repeatedly. This run issued 200 Page.navigate commands per binary.

Common ancestor 1e8f951ef2 → HEAD 62e8bec524; benchmark order: head-first; workflow: success.

Session resilience

Metric	Base	HEAD
Attempted / planned	200 / 200	200 / 200
Direct observable pass	197 / 200	192 / 200
Failures / recovered	3 / 3	8 / 8
Recovery failures	0	0
Lifecycle order violations	0	0
Network order violations	0	0
Wall time	675.7 s	840.8 s

A public-page failure is reported separately from an unrecoverable session. The soak requires all navigation attempts, zero failed recovery, zero lifecycle/network ordering violations, and complete resource evidence.

Process-tree memory

Metric	Base	HEAD	Δ
Boundary samples	200	200	0
Boundary RSS samples	200	200	0
Boundary PSS samples	200	200	0
Periodic peak RSS	268.8 MiB	297.8 MiB	+29.1 MiB
Periodic peak PSS	266.5 MiB	294.3 MiB	+27.7 MiB
RSS · first 10 avg	183.9 MiB	174.5 MiB	-9.4 MiB
RSS · last 10 avg	194.0 MiB	192.1 MiB	-1.9 MiB
RSS · last − first window	+10.2 MiB	+17.7 MiB	+7.5 MiB
RSS · warm slope / 100 nav	+5.4 MiB	+6.7 MiB	+1.2 MiB
PSS · first 10 avg	180.5 MiB	170.9 MiB	-9.6 MiB
PSS · last 10 avg	191.6 MiB	188.6 MiB	-3.1 MiB
PSS · last − first window	+11.1 MiB	+17.7 MiB	+6.6 MiB
PSS · warm slope / 100 nav	+6.3 MiB	+6.7 MiB	+0.5 MiB
Peak file descriptors	57	58	1
Peak threads	29	29	0

HEAD memory by 50-navigation quarter

HEAD range	Avg RSS	Final RSS	Peak RSS	Avg PSS	Final PSS	Peak FD
1–50	177.1 MiB	158.1 MiB	231.5 MiB	173.4 MiB	154.4 MiB	51
51–100	183.5 MiB	177.9 MiB	235.3 MiB	179.8 MiB	174.4 MiB	55
101–150	187.3 MiB	174.5 MiB	238.1 MiB	183.6 MiB	170.3 MiB	55
151–200	191.5 MiB	186.4 MiB	257.2 MiB	187.8 MiB	183.0 MiB	54

Raw reports and per-navigation boundary samples: workflow run and full sequential-navigation-soak-results artifact.

Public-site timing and memory are observational. Use the A/B deltas and trend shape as evidence, not as a deterministic performance threshold.

[lanyue-llk]

Closing this to keep the Web Fetch correction effort scoped to the consolidated renderer fixes in #1501. The PermissionStatus/EventTarget change is a valid independent standards improvement, but it is unrelated to the measured Web Fetch regressions and is not required by the current evidence. It can be reopened or resubmitted when that product behavior is independently prioritized.