refactor: OneCycleLR replaces SequentialLR(LinearLR, CosineAnnealingLR)

One scheduler object does warmup + cosine relaxation; pct_start=warmup_frac
is the explicit warmup. cycle_momentum=False so it doesn't clobber the
configured AdamW betas (adam_beta1). Curve (100 steps, 20% warmup, 3e-4):
peaks 3e-4 at step ~19 via smooth cos ramp (vs old linear), 1.7e-4 at step
10 where 5e-4 had diverged. Smoke + all verify gates green.

Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>

src/vgrout/train.py

@@ -413,17 +413,13 @@ def main(cfg: Config) -> int:
     # ── optimizer + schedule ── (A and B of both blocks; masks route grads)
     opt = torch.optim.AdamW(
         delta_params, lr=lr, weight_decay=cfg.weight_decay, betas=(adam_beta1, adam_beta2))
-    # Fractional warmup preserves the intended schedule across preset lengths.
-    warmup_steps = max(1, int(cfg.warmup_frac * steps))
-    sched = torch.optim.lr_scheduler.SequentialLR(
-        opt,
-        schedulers=[
-            torch.optim.lr_scheduler.LinearLR(opt, start_factor=1e-3, end_factor=1.0,
-                                              total_iters=warmup_steps),
-            torch.optim.lr_scheduler.CosineAnnealingLR(opt, T_max=max(1, steps - warmup_steps)),
-        ],
-        milestones=[warmup_steps],
-    )
+    # OneCycle does warmup + cosine relaxation in one object: cos ramp from lr/div_factor
+    # up to lr over the first pct_start of steps (the explicit warmup), then cos anneal to
+    # ~0. cycle_momentum=False so it leaves the configured AdamW betas alone (else it would
+    # clobber adam_beta1). pct_start = warmup_frac keeps warmup fractional across presets.
+    sched = torch.optim.lr_scheduler.OneCycleLR(
+        opt, max_lr=lr, total_steps=steps, pct_start=cfg.warmup_frac,
+        anneal_strategy="cos", div_factor=25.0, final_div_factor=1e4, cycle_momentum=False)
 
     # ── generation config ──
     # Use the same sampling policy for training and evaluation.