Fix incomplete activation validation in HausdorffDTLoss (Project-MONAI#8841)

The validation check for mutually exclusive activation options was
incomplete - it only checked sigmoid and softmax but not other_act,
despite the error message explicitly mentioning other_act.

Without this check, passing e.g. `sigmoid=True, other_act=relu` silently
stacks both activations in `forward()` (`other_act(sigmoid(x))`) instead
of applying only one, producing an incorrect loss with no warning.

Before:
```python
if int(sigmoid) + int(softmax) > 1:
    raise ValueError("... [sigmoid=True, softmax=True, other_act is not None].")
```

After:
```python
if int(sigmoid) + int(softmax) + int(other_act is not None) > 1:
    raise ValueError("... [sigmoid=True, softmax=True, other_act is not None].")
```

This is consistent with the validation in dice.py and tversky.py which
correctly include all three options in the check.

Added tests for:
- sigmoid=True with other_act
- softmax=True with other_act
- All three options combined

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

Signed-off-by: Soumya Snigdha Kundu <soumya_snigdha.kundu@kcl.ac.uk>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>

Author: aymuos15

monai/losses/hausdorff_loss.py

@@ -83,7 +83,7 @@ def __init__(
         super().__init__(reduction=LossReduction(reduction).value)
         if other_act is not None and not callable(other_act):
             raise TypeError(f"other_act must be None or callable but is {type(other_act).__name__}.")
-        if int(sigmoid) + int(softmax) > 1:
+        if int(sigmoid) + int(softmax) + int(other_act is not None) > 1:
             raise ValueError("Incompatible values: more than 1 of [sigmoid=True, softmax=True, other_act is not None].")
 
         self.alpha = alpha

tests/losses/test_hausdorff_loss.py

@@ -212,6 +212,12 @@ def test_ill_shape(self):
     def test_ill_opts(self):
         with self.assertRaisesRegex(ValueError, ""):
             HausdorffDTLoss(sigmoid=True, softmax=True)
+        with self.assertRaisesRegex(ValueError, ""):
+            HausdorffDTLoss(sigmoid=True, other_act=torch.tanh)
+        with self.assertRaisesRegex(ValueError, ""):
+            HausdorffDTLoss(softmax=True, other_act=torch.tanh)
+        with self.assertRaisesRegex(ValueError, ""):
+            HausdorffDTLoss(sigmoid=True, softmax=True, other_act=torch.tanh)
         chn_input = torch.ones((1, 1, 3))
         chn_target = torch.ones((1, 1, 3))
         with self.assertRaisesRegex(ValueError, ""):
@@ -244,6 +250,12 @@ def test_ill_shape(self):
     def test_ill_opts(self):
         with self.assertRaisesRegex(ValueError, ""):
             LogHausdorffDTLoss(sigmoid=True, softmax=True)
+        with self.assertRaisesRegex(ValueError, ""):
+            LogHausdorffDTLoss(sigmoid=True, other_act=torch.tanh)
+        with self.assertRaisesRegex(ValueError, ""):
+            LogHausdorffDTLoss(softmax=True, other_act=torch.tanh)
+        with self.assertRaisesRegex(ValueError, ""):
+            LogHausdorffDTLoss(sigmoid=True, softmax=True, other_act=torch.tanh)
         chn_input = torch.ones((1, 1, 3))
         chn_target = torch.ones((1, 1, 3))
         with self.assertRaisesRegex(ValueError, ""):