[BUG] HPO presets fail in multimodal

**Bug Report Checklist**

- [x] I provided code that demonstrates a minimal reproducible example.
- [ ] I confirmed bug exists on the latest mainline of AutoGluon via source install.
- [x] I confirmed bug exists on the latest stable version of AutoGluon.

**Describe the bug**
HPO presets fail with `FileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmp5dil14go/ae09e279/config.yaml'`, i tried `high_quality_hpo` and `medium_quality_hpo`

**To Reproduce**
```
from autogluon.multimodal import MultiModalPredictor
import tempfile

with tempfile.TemporaryDirectory() as temp_dir:
predictor = MultiModalPredictor(label="label", presets="high_quality_hpo", problem_type="multiclass", eval_metric="acc", path=temp_dir)
predictor.fit(train_df, time_limit=180, hyperparameters={"model.hf_text.checkpoint_name": "microsoft/deberta-v3-small"})
predictions = predictor.predict(test_df)
return predictions
```

**Screenshots / Logs**
```
Current best trial: ae09e279 with val_accuracy=0.26907894015312195 and params={'model.names': ('hf_text', 'fusion_mlp'), 'optim.lr': 0.0007805190045868186, 'optim.optim_type': 'adamw', 'optim.max_epochs': 13, 'env.batch_size': 256, 'env.per_gpu_batch_size': 2, 'model.hf_text.checkpoint_name': 'microsoft/deberta-v3-small'}
╭───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Trial name status optim.lr optim.optim_type optim.max_epochs env.batch_size iter total time (s) val_accuracy │
├───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┤
│ ae09e279 TERMINATED 0.000780519 adamw 13 256 1 144.601 0.269079 │
│ 3e29dc2d TERMINATED 2.2048e-05 sgd 30 256 │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯

Removing non-optimal trials and only keep the best one.
Traceback (most recent call last):
File "/home/voorhs/repos/AutoIntent-experiments/automl_frameworks.py", line 312, in
main()
File "/home/voorhs/repos/AutoIntent-experiments/automl_frameworks.py", line 249, in main
predictions = evaluate_gluon(train_df, test_df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/voorhs/repos/AutoIntent-experiments/automl_frameworks.py", line 190, in evaluate_gluon
predictor.fit(train_df, time_limit=180, hyperparameters={"model.hf_text.checkpoint_name": "microsoft/deberta-v3-small"})
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/predictor.py", line 540, in fit
self._learner.fit(
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/learners/base.py", line 665, in fit
fit_returns = self.execute_fit()
^^^^^^^^^^^^^^^^^^
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/learners/base.py", line 570, in execute_fit
hyperparameter_tune(
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/utils/hpo.py", line 235, in hyperparameter_tune
learner = build_final_learner(
^^^^^^^^^^^^^^^^^^^^
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/utils/hpo.py", line 118, in build_final_learner
learner = BaseLearner._load_metadata(learner=learner, path=best_trial_path)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/autogluon/multimodal/learners/base.py", line 2270, in _load_metadata
config = OmegaConf.load(os.path.join(path, "config.yaml"))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/voorhs/repos/AutoIntent-experiments/.venv/lib/python3.12/site-packages/omegaconf/omegaconf.py", line 189, in load
with io.open(os.path.abspath(file_), "r", encoding="utf-8") as f:
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmp5dil14go/ae09e279/config.yaml'
```

**Installed Versions**

```python
INSTALLED VERSIONS
------------------
date : 2025-05-22
time : 01:05:51.058434
python : 3.12.3.final.0
OS : Linux
OS-release : 6.11.0-25-generic
Version : #25~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:20:50 UTC 2
machine : x86_64
processor : x86_64
num_cores : 16
cpu_ram_mb : 15386.4609375
cuda version : 12.550.144.03
num_gpus : 1
gpu_ram_mb : [5764]
avail_disk_size_mb : 86936

accelerate : 1.6.0
autogluon : 1.3.0
autogluon.common : 1.3.0
autogluon.core : 1.3.0
autogluon.features : 1.3.0
autogluon.multimodal : 1.3.0
autogluon.tabular : 1.3.0
autogluon.timeseries : 1.3.0
boto3 : 1.38.21
catboost : 1.2.8
coreforecast : 0.0.15
defusedxml : 0.7.1
einops : 0.8.1
evaluate : 0.4.3
fastai : 2.8.1
fugue : 0.9.1
gluonts : 0.16.1
huggingface-hub : 0.31.4
hyperopt : 0.2.7
imodels : None
jinja2 : 3.1.6
joblib : 1.4.2
jsonschema : 4.23.0
lightgbm : 4.6.0
lightning : 2.5.1.post0
matplotlib : 3.10.3
mlforecast : 0.13.6
networkx : 3.2.1
nlpaug : 1.1.11
nltk : 3.9.1
numpy : 1.26.4
nvidia-ml-py3 : 7.352.0
omegaconf : 2.3.0
onnx : None
onnxruntime : None
onnxruntime-gpu : None
openmim : 0.3.9
optimum : None
optimum-intel : None
orjson : 3.10.18
pandas : 2.2.3
pdf2image : 1.17.0
Pillow : 11.2.1
psutil : 7.0.0
pyarrow : 17.0.0
pytesseract : 0.3.13
pytorch-lightning : 2.5.1.post0
pytorch-metric-learning: 2.8.1
ray : 2.44.1
requests : 2.32.3
scikit-image : 0.25.2
scikit-learn : 1.5.2
scikit-learn-intelex : None
scipy : 1.12.0
seqeval : 1.2.2
skl2onnx : None
spacy : 3.8.6
statsforecast : 2.0.1
tabpfn : None
tensorboard : 2.19.0
text-unidecode : 1.3
timm : 1.0.3
torch : 2.6.0
torchmetrics : 1.7.1
torchvision : 0.21.0
tqdm : 4.66.6
transformers : 4.49.0
utilsforecast : 0.2.10
xgboost : 2.1.4
```