Fix TypeError when simpletuner.__file__ is None during training launch

[Copilot]

simpletuner.__file__ can be None under certain install configurations (namespace packages, some editable installs), causing Path(None) to raise TypeError at training start.

train_py = Path(simpletuner.__file__).parent / "train.py"
           ~~~~^^^^^^^^^^^^^^^^^^^^^^
TypeError: argument should be a str or an os.PathLike object where __fspath__ returns a str, not 'NoneType'

Changes

• simpletuner/__init__.py — Add _get_package_dir() helper that resolves the package directory via __file__, falling back to importlib.util.find_spec (checking both spec.origin and spec.submodule_search_locations)
• trainer.py, cli/train.py, test_e2e_examples.py — Replace all bare Path(simpletuner.__file__).parent calls with simpletuner._get_package_dir()
• paths.py — Remove duplicated __file__ is None guard in get_simpletuner_root(), delegate to the shared helper
• tests/test_get_package_dir.py — Tests covering normal path, __file__ = None with importlib fallback, and unresolvable package cases

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Training start immediately fails with TypeError</issue_title>
<issue_description>Hi there, thanks for your work on this project. I have been able to run the Web UI in a venv in Linux + ROCm with success, but I get this error always a few seconds after I try to start training:

Multiple distributions found for package optimum. Picked distribution: optimum-quanto
[RANK 0] 2026-02-08 11:58:31,628 [INFO] Patched Attention with flexible fusion (permanent=True)
[RANK 0] 2026-02-08 11:58:31,704 [ERROR] Function error: argument should be a str or an os.PathLike object where __fspath__ returns a str, not 'NoneType'
[RANK 0] 2026-02-08 11:58:31,705 [ERROR] Traceback (most recent call last):
  File "<string>", line 322, in <module>
    result = target_func(wrapped_config)
  File "/workspace/SimpleTuner/simpletuner/helpers/training/trainer.py", line 6780, in run_trainer_job
    accelerate_result = _launch_with_accelerate()
  File "/workspace/SimpleTuner/simpletuner/helpers/training/trainer.py", line 6606, in _launch_with_accelerate
    train_py = Path(simpletuner.__file__).parent / "train.py"
               ~~~~^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib64/python3.13/pathlib/_local.py", line 503, in __init__
    super().__init__(*args)
    ~~~~~~~~~~~~~~~~^^^^^^^
  File "/usr/lib64/python3.13/pathlib/_local.py", line 132, in __init__
    raise TypeError(
    ...<2 lines>...
        f"not {type(path).__name__!r}")
TypeError: argument should be a str or an os.PathLike object where __fspath__ returns a str, not 'NoneType'

& Just in case you need the info:

(venv) [root@517da84a37fa workspace]# rocminfo
ROCk module is loaded
=====================    
HSA System Attributes    
=====================    
Runtime Version:         1.18
Runtime Ext Version:     1.16
System Timestamp Freq.:  1000.000000MHz
Sig. Max Wait Duration:  18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model:           LARGE                              
System Endianness:       LITTLE                             
Mwaitx:                  DISABLED
XNACK enabled:           NO
DMAbuf Support:          YES
VMM Support:             YES

==========               
HSA Agents               
==========               
*******                  
Agent 1                  
*******                  
  Name:                    AMD RYZEN AI MAX+ 395 w/ Radeon 8060S
  Uuid:                    CPU-XX                             
  Marketing Name:          AMD RYZEN AI MAX+ 395 w/ Radeon 8060S
  Vendor Name:             CPU                                
  Feature:                 None specified                     
  Profile:                 FULL_PROFILE                       
  Float Round Mode:        NEAR                               
  Max Queue Number:        0(0x0)                             
  Queue Min Size:          0(0x0)                             
  Queue Max Size:          0(0x0)                             
  Queue Type:              MULTI                              
  Node:                    0                                  
  Device Type:             CPU                                
  Cache Info:              
    L1:                      49152(0xc000) KB                   
  Chip ID:                 0(0x0)                             
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   5187                               
  BDFID:                   0                                  
  Internal Node ID:        0                                  
  Compute Unit:            32                                 
  SIMDs per CU:            0                                  
  Shader Engines:          0                                  
  Shader Arrs. per Eng.:   0                                  
  WatchPts on Addr. Ranges:1                                  
  Memory Properties:       
  Features:                None
  Pool Info:               
    Pool 1                   
      Segment:                 GLOBAL; FLAGS: FINE GRAINED        
      Size:                    131155396(0x7d145c4) KB            
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Recommended Granule:4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 2                   
      Segment:                 GLOBAL; FLAGS: EXTENDED FINE GRAINED
      Size:                    131155396(0x7d145c4) KB            
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Recommended Granule:4KB                                
      Alloc Alignment:         4KB       ...

</details>



<!-- START COPILOT CODING AGENT SUFFIX -->

- Fixes bghira/SimpleTuner#2581

<!-- START COPILOT CODING AGENT TIPS -->
---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.