Fix minor bug on `PolynomialLR` invocation (#6436)

Resolves issue reported at https://github.com/pytorch/vision/commit/6e535db255cee3ce878dd7a54dda01d4ec8932c1#commitcomment-81409388

There seems to be a misspelling on the name of the parameter. This PR updates `total_steps` to `total_iters` which is the correct argument.

references/segmentation/train.py

@@ -186,7 +186,7 @@ def main(args):
 
     iters_per_epoch = len(data_loader)
     main_lr_scheduler = PolynomialLR(
-        optimizer, total_steps=iters_per_epoch * (args.epochs - args.lr_warmup_epochs), power=0.9
+        optimizer, total_iters=iters_per_epoch * (args.epochs - args.lr_warmup_epochs), power=0.9
     )
 
     if args.lr_warmup_epochs > 0: