Suunto Nautic: upstream the driver to libdivecomputer suunto-nautic branch

[latishab]

Jef Driesen started an official suunto-nautic branch (early skeleton, device_open/foreach are TODO, wired for serial/HDLC not BLE). Our fork is far ahead with a working BLE driver + parser. Contribute the working implementation upstream with clean, trailer-free commits. Includes porting the reference tests to C (dctool) and fixing any stale header/parser doc comments before submitting.

[urbamax]

@latishab — three fixes have landed in deepsealabs/libdivecomputer since the base of libdivecomputer#73 and aren't on that PR's branch yet:

• 6684375 suunto_nautic: 45-byte gas-record stride (Suunto Nautic: wire multi-gas (gas switches + gas mixes) #33) — #73 still has i * 4 and the "4 bytes each" comment.
• c4ae840 skip foreign 0x07 frames on a contended link.
• c059571 begin pressure = max reading (transmitter dropout).

Also: libdivecomputer#75 (mine) adds DC_FIELD_PPO2_MAX reading Gas[0]+5, stacked on #73's stride-4 base. In Max's app submodule I've already folded it onto your 45-byte version. Happy to rebase #75 once the fix reaches #73, or you can fold the field into the multigas commit.

[urbamax]

Bumping this — Jef's review of libdivecomputer#73 is now active (he asked for device logs, I've attached a full 2-dive download log to the PR), so the branch he's reviewing should be the complete one.

Four things are validated but not on deepsealabs:suunto-nautic (2a1ca0b):

• 6684375 — 45-byte gas-record stride (#33). #73 still has i * 4 + the "4 bytes each" comment; this is the most visible one in review.
• c4ae840 — skip foreign 0x07 frames on a contended link.
• c059571 — begin pressure = max reading (transmitter dropout).
• new: deepsealabs/libdivecomputer-upstream#4 — STREAM_STOP (0x11) + stream re-arm (0x0B on 0x240E) after each /Data. Without it a dc_device_foreach() on a multi-dive watch only returns the first dive (423 Locked, then a buffer replay). Verified on-device: two dives now download as distinct profiles. This is the trustthegoose "only first dive" case on #29.

If you cherry-pick #4 into suunto-nautic it drops straight in (I confirmed against the tip). Could bundle all four in one pass so #73's branch matches what's shipping in the fork.

#75 (DC_FIELD_PPO2_MAX) still stacked on the stride-4 base — happy to rebase it once the stride fix reaches #73.