[Platform][Linux] g_source_remove() is used on already removed source ID

[arkq]

Reproduction steps

1. Start `chip-lighting-app --ble-device 0` on one device
2. Run `chip-tool pairing ble-wifi 0x01 WIFI PASSW 20202021 3840` on other device which does not support all required BLE features (on lighting-app device, in Wireshark `bthci_evt.le_features` is all zero)
3. lighting-app will print:


[1663922441.124142][1652598:1652616] CHIP:DL: c1 BluezCharacteristicWriteFD mtu, 517
[1663922441.124182][1652598:1652616] CHIP:DL: FAIL: short read in BluezCharacteristicWriteFD (0)     <- source callback will return FALSE
[1663922441.222954][1652598:1652616] CHIP:DL: CHIPoBLE subscribe received
[1663922441.223036][1652598:1652598] CHIP:BLE: no endpoint for sub recvd
[1663922441.223061][1652598:1652598] CHIP:DL: HandlePlatformSpecificBLEEvent 32775
[1663922441.223076][1652598:1652598] CHIP:DL: Receive kCHIPoBLEConnectionEstablished
[1663922458.174623][1652598:1652616] CHIP:DL: Bluez disconnected
[1663922458.174675][1652598:1652616] CHIP:DL: Bluez notify CHIPoBluez connection disconnected

(process:1652598): GLib-CRITICAL **: 10:40:58.174: Source ID 40 was not found when attempting to remove it       <- calling g_source_remove() on removed source

### Platform

other

### Platform Version(s)

_No response_

### Type

Manually tested with SDK

### (Optional) If manually tested please explain why this is only manually tested

No automatic test case for such scenario.

### Anything else?

_No response_