Feat: add devstral model support (#2880) [skip ci]

* fix: do not add training and training_detail block by default

* fixed: magistral docs

* fix: address pad adding new fields and use built-in from_openai

* feat: try enable multiprocessing

* fix: check for keys before deleting attn_mask

* feat: add mistral pad test

* feat: add tool calling test

* feat: add devstral tokenizer tests

* fix: comma format

* chore: remove unused support_preprocessing as tokenizer is pickable now

* chore: update magistral doc

* feat: add devstral readme and example

* chore: refactor error handling

examples/devstral/devstral-small-qlora.yml

@@ -0,0 +1,64 @@
+base_model: mistralai/Devstral-Small-2505
+
+# Automatically upload checkpoint and final model to HF
+# hub_model_id: username/custom_model_name
+
+# Enable to use mistral-common tokenizer
+tokenizer_use_mistral_common: true
+
+load_in_8bit: false
+load_in_4bit: true
+
+plugins:
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+
+datasets:
+  - path: fozziethebeat/alpaca_messages_2k_test
+    type: chat_template
+
+dataset_prepared_path: last_run_prepared
+val_set_size: 0.1
+output_dir: ./outputs/qlora-out
+
+adapter: qlora
+lora_model_dir:
+
+sequence_len: 2048
+sample_packing: true
+pad_to_sequence_len: true
+
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0
+lora_target_linear: true
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 4
+micro_batch_size: 2
+num_epochs: 1
+optimizer: adamw_torch
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+bf16: auto
+tf32: false
+
+gradient_checkpointing: true
+resume_from_checkpoint:
+logging_steps: 1
+flash_attention: true
+
+loss_watchdog_threshold: 5.0
+loss_watchdog_patience: 3
+
+warmup_ratio: 0.05
+evals_per_epoch: 4
+saves_per_epoch: 1
+
+weight_decay: 0.0
+special_tokens: