catalog(fish-s2): add Imogen_Contralto to voice dropdown

Imogen pitch-shifted -3 semitones (rubberband, formants follow) -> deeper
contralto register. Staged + verified cloning; added to reference_id options
(33 voices). version 4->5.

docs/asset-engine/services.yaml

@@ -619,7 +619,7 @@ services:
       Trained 10M+ hours, dual-AR. Released March 2026. Heavy: ~240s compile
       warmup on cold start, ~realtime throughput once warm.
     category: tts
-    version: 4
+    version: 5
     host: irv-ml1
     lifecycle:
       stack: fish-s2
@@ -717,6 +717,7 @@ services:
           - Beatrice
           - Eleanor
           - Imogen
+          - Imogen_Contralto
           - Connor
           - Cora
           - Elena
@@ -742,12 +743,14 @@ services:
           - Thomas
         description: >
           Voice = a staged clone reference picked by name (THE working voice
-          path on this build; verified live 2026-06-01). 32 voices staged in
+          path on this build; verified live 2026-06-01). 33 voices staged in
           /worktank/fish-s2/references/: 28 from the dia library + glados + 3
-          British-female VCTK voices (Beatrice/Eleanor/Imogen). Default Emily.
-          Subtle British (Southern-England) female accents: Imogen (VCTK p225),
-          Eleanor (p228), Beatrice (p229) — consenting VCTK volunteers (CC BY
-          4.0), NOT modeled on or representing any public figure. Other female:
+          British-female VCTK voices (Beatrice/Eleanor/Imogen) + Imogen_Contralto.
+          Default Emily. Subtle British (Southern-England) female accents: Imogen
+          (VCTK p225), Eleanor (p228), Beatrice (p229) — consenting VCTK
+          volunteers (CC BY 4.0), NOT modeled on or representing any public
+          figure. Imogen_Contralto = Imogen pitch-shifted -3 semitones
+          (rubberband, formants follow) for a deeper contralto register. Other female:
           Abigail, Alice, Cora, Elena, Emily, Gianna, Jade, Layla, Olivia,
           glados. Resolves to <name>.wav + optional <name>.txt transcript;
           blank = model default/random speaker. To add: drop a clean 5–15s WAV