Added support for Corsair Virtuoso XT

[Anshuman01000001]

Adds support for the Corsair Virtuoso XT — both the wireless receiver (0x0a64) and wired USB (0x0a62).

I reverse-engineered the protocol via hidraw probing on Linux. The Virtuoso XT doesn't use the same protocol as CorsairVoidRich or CorsairVoidV2W — it communicates on interface 3 (Usage-Page 0xff42) and responds to a simple 0x02 0x00 battery request:

• Byte 0: report ID (0x01)
• Byte 1: status (0xf0 = connected, 0x00 = offline)
• Byte 2: battery percentage (0-100)

Charging state doesn't appear to be reported over HID. Tested on Arch Linux with the wireless receiver — battery percentage matches the headset's LED indicator.

Capabilities: battery status only for now.

[claude]

Code review

Bug: Battery level not validated against 0-100

File: `lib/devices/corsair_virtuoso_xt.hpp`, lines 98-109

`response[2]` is a `uint8_t` (range 0-255), but `BatteryResult::level_percent` is documented as 0-100 in result_types.hpp. A malformed response could yield 255, which also feeds `time_to_empty_min` with a nonsensical result. Multiple peer devices enforce this bound — `steelseries_arctis_nova_7.hpp` clamps with `if (level > 100) level = 100;`, and `logitech_gpro_x2_lightspeed.hpp` returns `protocolError` if out of range.

Suggested fix after line 98:

```cpp
if (battery_level > 100) {
return DeviceError::protocolError(
std::format("Battery percentage out of range: {}", battery_level));
}
```

---

CLAUDE.md: Class docblock embeds wire-format table (WHAT not WHY)

File: `lib/devices/corsair_virtuoso_xt.hpp`, lines 13-31 —

HeadsetControl/lib/devices/corsair_virtuoso_xt.hpp

Lines 12 to 32 in 3291423

	/**
	* @brief Corsair Virtuoso XT (Wireless + Wired)
	*
	* Protocol reverse-engineered via hidraw probing on Linux.
	*
	* Battery request: send 0x02 0x00 on interface 3 (Usage-Page 0xff42)
	* Response format (64 bytes):
	* [0] = 0x01 (report ID)
	* [1] = status flags (0xf0 = normal, TBD for charging)
	* [2] = battery percentage (0-100)
	* [3+] = zeros (unused)
	*
	* Volume events are broadcast unsolicited:
	* [0] = 0x0E
	* [1] = 0x00 (down), 0x01 (up), 0x02 (fast up)
	*
	* Wireless Product ID: 0x0a64 (receiver)
	* Wired Product ID: 0x0a62
	*/
	class CorsairVirtuosoXT : public CorsairDevice {

The `/** @brief */` block embeds a full byte-offset/hex-value protocol table and lists product IDs already in `SUPPORTED_PRODUCT_IDS`. CLAUDE.md says: "Default to writing no comments. Only add one when the WHY is non-obvious." Compared with `corsair_void_v2w.hpp`, whose docblock only documents non-obvious behavioral quirks, the wire-format table belongs in docs/ or the PR description. Keep only the reverse-engineering provenance note.

---

CLAUDE.md: Machine-specific device path in committed source

File: `lib/devices/corsair_virtuoso_xt.hpp`, line 70 —

HeadsetControl/lib/devices/corsair_virtuoso_xt.hpp

Lines 69 to 71 in 3291423

	// Send battery status request: 0x02 0x00
	// Discovered via hidraw probing on /dev/hidraw11 (interface 3, 0xff42)
	std::array<uint8_t, 2> request { 0x02, 0x00 };

The comment `// Discovered via hidraw probing on /dev/hidraw11 (interface 3, 0xff42)` embeds a machine-specific device node from the author's session. CLAUDE.md says: "Don't reference the current task, fix, or callers... since those belong in the PR description and rot as the codebase evolves." The interface (3) and usage-page (0xff42) are already in `getCapabilityDetail`; remove this line.

[Sapd]

Ignore the review about the comments. However the first bug might be relevant.

[Anshuman01000001]

I'll make the required changes. I will also try and adding more features for this headest after resolving this PR, Thanks.