[training_utils] fix: Update metrics.py to change calculate_log_prob_diff to calculate_diff

[victordion]

What does this PR do?

The function name calculate_log_prob_diff is misleading as the tensor it processes is not log probs but probs.

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: ...
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, veomni, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data, cfg, reward, fully_async, one_step_off
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

N/A

API and Usage Example

N/A

Design & Code Changes

N/A

[gemini-code-assist]

Code Review

This pull request renames the calculate_log_prob_diff function to calculate_diff and updates its implementation and call site to operate on probabilities instead of log-probabilities. The review feedback recommends reverting these changes to use log-probabilities, as they provide more sensitive debugging metrics for LLMs, and suggests an optimized implementation to reduce memory overhead during the calculation.