Client: dual-channel label-merge (mic_file + system_file)

The backend shipped dual-channel mode; wire the client to it. We already capture
mic (you) and system (others) separately, so send them as two files instead of the
mono mix — fixing the misattribution at the source.

- SparkControlClient: labelMergeDual(mic_file, system_file, self_name, self_vad);
  multipart generalized to N files; shared POST/retry/decode extracted.
- SessionPackager.rebasedSelfVadData: chunk-local [{start,end}] for self_vad;
  sliceAudio reused for both tracks.
- TranscriptPipeline.process: dual-channel chunking (slice mic+system, rebase
  timeline + self_vad per chunk) when system audio is healthy; mono mixed-file
  fallback (self folded into the timeline) otherwise.
- VisualCapture.finish: write the full visual_timeline.json (remote + self merged)
  but return REMOTE (vision) segments only — self travels via the mic channel.
- TranscriptAssembler: rank mic_channel highest (the user's own track wins).
- VoiceprintStore: store the clean mic_channel self voiceprint.
- SessionController: pass mic/system URLs + remote timeline + channel self-spans +
  self_name + systemHealthy; self_vad.json now reflects the channel-verified spans.

Validated END-TO-END against the live backend on the real misattributing session:
'Go Bitcoin' (remote) is now attributed to Unknown_0, NOT the user; the user's own
lines come back source=mic_channel; per-channel ASR recovered fuller remote text.
36/36 XCTest (4 new: self_vad rebase, mic_channel ranking + voiceprint storage).

Ten31TranscriptsTests/Phase5Tests.swift

@@ -29,6 +29,31 @@ final class Phase5Tests: XCTestCase {
         XCTAssertEqual(arr[1]["end"] as? Double, 110)
     }
 
+    func testRebaseSelfVadClipsAndRebases() throws {
+        let spans = [VADSpan(start: 140, end: 160, confidence: 0.9),
+                     VADSpan(start: 200, end: 260, confidence: 0.8)]
+        let data = try SessionPackager.rebasedSelfVadData(spans, start: 150, end: 300)
+        let arr = try XCTUnwrap(JSONSerialization.jsonObject(with: data) as? [[String: Any]])
+        XCTAssertEqual(arr.count, 2)
+        XCTAssertEqual(arr[0]["start"] as? Double, 0)
+        XCTAssertEqual(arr[0]["end"] as? Double, 10)     // 160 clipped at 150 base → 0–10
+        XCTAssertEqual(arr[1]["start"] as? Double, 50)
+        XCTAssertEqual(arr[1]["end"] as? Double, 110)
+        XCTAssertNil(arr[0]["name"])                      // self_vad carries no name
+    }
+
+    func testAssembleRanksMicChannelOverVisual() throws {
+        // Same person resolved by visual in one chunk and by the mic channel in
+        // another → the mic-channel attribution (the user's own track) wins.
+        let visual = #"{"duration":100,"speakers":[{"cluster":"Speaker_0","name":"Grant","source":"visual","overlap_confidence":0.99,"fingerprint":[0.1]}],"segments":[],"fingerprints":{},"models":{}}"#
+        let mic = #"{"duration":100,"speakers":[{"cluster":"mic","name":"Grant","source":"mic_channel","fingerprint":[0.2]}],"segments":[],"fingerprints":{"Grant":[0.2]},"models":{}}"#
+        let rv = try JSONDecoder().decode(LabelMergeResponse.self, from: Data(visual.utf8))
+        let rm = try JSONDecoder().decode(LabelMergeResponse.self, from: Data(mic.utf8))
+        let asm = TranscriptAssembler.assemble(sessionId: "s", app: "meet",
+            chunks: [.init(chunkStart: 0, response: rv), .init(chunkStart: 100, response: rm)])
+        XCTAssertEqual(asm.speakersFile.speakers.first { $0.name == "Grant" }?.source, "mic_channel")
+    }
+
     func testAssembleOffsetsAndUnifies() throws {
         let resp0 = #"{"duration":150,"speakers":[{"cluster":"Speaker_0","name":"Grant","source":"visual","overlap_confidence":0.99,"fingerprint":[0.1,0.2]}],"segments":[{"start_ms":1000,"end_ms":2000,"speaker":"Grant","text":"hi"}],"fingerprints":{"Grant":[0.1,0.2]},"models":{"diarization":"x"}}"#
         let resp1 = #"{"duration":100,"speakers":[{"cluster":"Speaker_0","name":"Sarah","source":"voiceprint","match_similarity":0.7,"fingerprint":[0.3,0.4]},{"cluster":"Speaker_1","name":"Unknown_0","source":"unmatched"}],"segments":[{"start_ms":500,"end_ms":1500,"speaker":"Sarah","text":"hello"}],"fingerprints":{"Sarah":[0.3,0.4]},"models":{"diarization":"x"}}"#