DPO support loss types (#3566)

* Support loss_type/loss_weights DPO

* Validate dpo loss type/weights only set for dpo

* Tests: Update ipo tests to use new path

* Docs: Update docs for new ipo path

* PR fixes - typo/validation

* PR nit - warning

* chore: fix warnings arg

---------

Co-authored-by: NanoCode012 <nano@axolotl.ai>

docs/agents/preference_tuning.md

@@ -38,7 +38,7 @@ No vLLM server needed (unlike GRPO). Offline RL with pre-collected preference da
 
 1. Paired preference data (chosen + rejected)?
    - Default → `rl: dpo`
-   - Overfitting → `rl: ipo`
+   - Overfitting → `rl: dpo, dpo_loss_type: ["ipo"]`
    - VRAM-limited → `rl: orpo` (no ref model)
    - Length-sensitive → `rl: simpo` (no ref model)
 2. Only binary labels (good/bad)? → `rl: kto`

docs/rlhf.qmd

@@ -320,8 +320,10 @@ The input format is a simple JSON input with customizable fields based on the ab
 As IPO is just DPO with a different loss function, all supported dataset formats for [DPO](#dpo) are also supported for IPO.
 
 ```yaml
-rl: ipo
+rl: dpo
+dpo_loss_type: ["ipo"]
 ```
+*Note:* Passing `rl: ipo` directly is still supported, but will soon be deprecated.
 
 ### ORPO