Fix wrong data-link verdicts on MagSafe and USB 2.0 cables

What's fixed

MagSafe ports no longer claim "Running at full data speed (40 Gbps)" (issue #195). MagSafe carries no data lanes, so any data-speed verdict on it was wrong. On every M-class MacBook, the MagSafe port and the first USB-C port share a controller die in macOS's view, and a port-identifier collision was letting WhatCable attribute the USB-C port's Thunderbolt lane state to MagSafe. Fixed across every Mac in the customer-probe set (M1 Pro through M5 Max). Thanks @blackketter for the report.
USB-C ports with a USB 2.0 cable no longer show contradictory speeds. The cable badge would correctly say "USB 2.0 (480 Mbps)" while the inline verdict said "Running at full data speed (40 Gbps)". The verdict was silently overriding the cable's own report whenever they disagreed; now the diagnostic reports both numbers honestly when there's no Thunderbolt controller cross-check available. Thanks @bigskookum for spotting it on a comment in #195.
Defence in depth. The data-link diagnostic now refuses to verdict on any port that doesn't advertise at least one data transport (USB2/USB3/USB4/CIO/DisplayPort). Belt and braces against a future regression re-introducing the same class of bug.