**Bug fix**
Fixed an issue that occurred when specifying a floating-point value for `offset` during testing.
**New**
Added new configuration parameters for controlling gradient flow calculations, as well as a parameter defining the IoU threshold used for metric computation.
**Bug fix**
1. Fixed an error in dataset sampling during automatic splitting.
2. Fixed a memory leak when loading images in formats other than `.tif`.
**Changes:**
1. `shuffle` can now be used with both `pre-split` and `split` methods.
2. `offsets` can be specified as floating-point values for dynamic computation.
3. `size` and `offsets` now support mixed formats (i.e., both `int` and `float`).
4. F1 and mAP metrics are now computed by default in `micro`, `macro`, and `per_class` variations.
5. Classes `BoundaryExclusion` and `IntensityDiversification` have been renamed to follow the MONAI naming style.
6. Minor updates to the README for clarity and consistency.
Replaced typing module generics like List, Dict, Tuple with built-in alternatives (list, dict, tuple).
Updated code to use new union syntax (X | Y) instead of Union[X, Y].
Added arguments for creating only prediction masks, as well as refusing to save the output;
Added execution time display;
The pretrained_weight parameter has been moved to common;
Parameters related to ensembles have been removed.
