fix: detect wired battery-less devices and resolve registry-absent images

[davidbudnick]

Context

• wired battery-less devices got dropped when the hid++ feature walk didn't complete
• now fall back to a cheap root.getFeature probe so they stay listed
• helps the wider class of wired g-series, not just one model — e.g. g402, g102, g303, g302, g513, g512
• alias registry-absent devices to a sibling depot's render so they aren't blank

Testing

• openlogi-hid + openlogi-gui tests and clippy pass
• added resolver tests incl. one guarding the alias from over-matching

[greptile-apps]

Greptile Summary

This PR fixes two issues with directly-connected (wired, battery-less) Logitech devices: a HID++ inventory path that dropped real peripherals when their feature-table walk failed, and a missing image for the G502 HERO in the GUI asset resolver.

• openlogi-hid: When probe_direct's full FeatureSet walk doesn't complete, a new peripheral_caps_via_root fallback issues individual root.getFeature round-trips for a fixed set of driving feature IDs; if any respond the device is admitted as a peripheral rather than silently dropped. PERIPHERAL_FEATURE_IDS is missing 0x8070 (ColorLedEffects), which Capabilities::from_feature_ids counts as a lighting capability \u2014 a keyboard with only that feature and a failing walk would still be dropped.
• openlogi-gui: A curated aliased_depot function maps the G502 HERO's PID (c08b) and codename substring to the g502_core render depot, while the resolve call-site overrides display_name with the device's own codename so it isn't relabelled "G502". Three new unit tests cover PID match, codename fallback, and the guard against over-matching.

Confidence Score: 3/5

Safe to merge for the GUI change; the HID fallback has a gap that could leave certain wired keyboards undetected.

The PERIPHERAL_FEATURE_IDS constant in inventory.rs omits 0x8070 (ColorLedEffects), a feature that Capabilities::from_feature_ids recognises as driving the lighting panel. A wired keyboard that advertises only 0x8070 and whose feature-table walk fails would pass through peripheral_caps_via_root with an empty hit list and be dropped from the device list — the same failure mode the PR intends to fix. The GUI alias logic is well-constructed and well-tested.

crates/openlogi-hid/src/inventory.rs — specifically the PERIPHERAL_FEATURE_IDS constant and its coverage of lighting feature IDs.

Important Files Changed

Filename	Overview
crates/openlogi-hid/src/inventory.rs	Adds peripheral_caps_via_root fallback for wired battery-less mice whose feature-table walk fails; PERIPHERAL_FEATURE_IDS is missing 0x8070 (ColorLedEffects), so a keyboard with only that lighting feature and a failing walk would still be dropped.
crates/openlogi-gui/src/asset/mod.rs	Adds curated PID/codename alias resolving G502 HERO to g502_core render with correct identity preservation; logic is sound and well-tested, minor style note on substring matching for codename aliases.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[probe_direct called] --> B[probe_or_reuse: full FeatureSet walk]
    B --> C{walk_succeeded?\nprobe.capabilities.is_some}
    C -- Yes --> D[effective_caps = probe.capabilities]
    C -- No --> E[peripheral_caps_via_root fallback\n8x root.getFeature round-trips]
    E --> F{any PERIPHERAL_FEATURE_IDS\nresponded?}
    F -- Yes --> G[fallback_caps = Some caps]
    F -- No --> H[fallback_caps = None]
    G --> I[effective_caps = fallback_caps]
    H --> J[effective_caps = None\ncaps = default]
    D --> K{is_peripheral?\nbattery OR buttons OR pointer OR lighting}
    I --> K
    J --> L[Rejected — not a peripheral\nhealthy = walk_succeeded = false\nreplay last inventory]
    K -- No --> L
    K -- Yes --> M[Admitted as PairedDevice\ncapabilities = effective_caps\nhealthy = true]

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
    A[probe_direct called] --> B[probe_or_reuse: full FeatureSet walk]
    B --> C{walk_succeeded?\nprobe.capabilities.is_some}
    C -- Yes --> D[effective_caps = probe.capabilities]
    C -- No --> E[peripheral_caps_via_root fallback\n8x root.getFeature round-trips]
    E --> F{any PERIPHERAL_FEATURE_IDS\nresponded?}
    F -- Yes --> G[fallback_caps = Some caps]
    F -- No --> H[fallback_caps = None]
    G --> I[effective_caps = fallback_caps]
    H --> J[effective_caps = None\ncaps = default]
    D --> K{is_peripheral?\nbattery OR buttons OR pointer OR lighting}
    I --> K
    J --> L[Rejected — not a peripheral\nhealthy = walk_succeeded = false\nreplay last inventory]
    K -- No --> L
    K -- Yes --> M[Admitted as PairedDevice\ncapabilities = effective_caps\nhealthy = true]

Loading

_{Reviews (1): Last reviewed commit: "fix(gui): alias the G502 HERO to the g50..." | Re-trigger Greptile}

[greptile-apps]

PERIPHERAL_FEATURE_IDS is missing 0x8070 (ColorLedEffects), which Capabilities::from_feature_ids recognises as a driving lighting feature alongside 0x8080. A wired keyboard that exposes only 0x8070 and whose feature-table walk fails would have an empty present list after the fallback probe, causing peripheral_caps_via_root to return None and the device to be dropped — the exact failure mode this PR is designed to fix.

Suggested change

	const PERIPHERAL_FEATURE_IDS: [u16; 8] = [
	0x1b00, 0x1b01, 0x1b02, 0x1b03, 0x1b04, 0x2201, 0x2202, 0x8080,
	];
	const PERIPHERAL_FEATURE_IDS: [u16; 9] = [
	0x1b00, 0x1b01, 0x1b02, 0x1b03, 0x1b04, 0x2201, 0x2202, 0x8070, 0x8080,
	];

[greptile-apps]

Codename alias uses substring match — name.contains(needle) means any future device whose OS product string happens to contain "g502 hero" (e.g. a hypothetical "G502 HERO SE" with a different chassis) would borrow the wrong render without any explicit entry. Since the PID path is authoritative and takes priority, the risk is low today, but as the alias table grows, an exact-match or anchored check would be safer and communicate intent more clearly.

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!