feat: add cosine LR decay schedule to trainer and scheduler

- Add lr_schedule param (constant|cosine) to SelvaLoraTrainer
- Cosine decays LR from initial value to ~0 after warmup, preventing
  the oscillation observed at steps 6000-8000 with lr=2e-4 flat
- Wire lr_schedule through scheduler _PARAM_DEFAULTS and _train_inner call
- Add g5_r128_lr_2e4_cosine and g5_r128_lr_3e4_cosine to r128_sweet_spot sweep

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

experiments/r128_sweet_spot.json

@@ -82,6 +82,21 @@
       "description": "Rank 256 + LR=3e-4. Best rank + best LR candidate combined.",
       "rank": 256,
       "lr": 3e-4
+    },
+
+    {
+      "id": "g5_r128_lr_2e4_cosine",
+      "group": "cosine",
+      "description": "LR=2e-4 + cosine decay. Fixes the oscillation observed at step 6000–8000 by decaying LR to ~0 instead of staying flat.",
+      "lr": 2e-4,
+      "lr_schedule": "cosine"
+    },
+    {
+      "id": "g5_r128_lr_3e4_cosine",
+      "group": "cosine",
+      "description": "LR=3e-4 + cosine decay. Higher LR with decay — should reach lower loss faster then lock in.",
+      "lr": 3e-4,
+      "lr_schedule": "cosine"
     }
 
   ]