Add PHYBench

[suencgo]

##Motivation
This PR adds support for the PHYBench dataset to the OpenCompass framework. PHYBench is a benchmark designed for evaluating large language models on symbolic physics problems with structured LaTeX answers. The goal is to enable high-fidelity evaluation of models' reasoning capabilities in physics using expression-level symbolic comparison.

##Modification
Added PhyBenchDataset for loading the dataset from a local JSON file.

Implemented a custom evaluator MathEEDEvaluator using the EED (Extended Edit Distance) metric for symbolic similarity.

Integrated three utility files: EED.py, extended_zss.py, and latex_pre_process.py, which are used by the evaluator to process and compare symbolic math expressions.

Registered the dataset and evaluator in phybench_gen.py under the configs/datasets/PHYBench directory.

Configured the dataset's metadata in datasets_info.py to support local loading.

Checklist

Before PR:

• Pre-commit or other linting tools are used to fix the potential lint issues.
• Bug fixes are fully covered by unit tests, the case that causes the bug should be added in the unit tests.
• The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
• The documentation has been modified accordingly, like docstring or example tutorials.

After PR:

• If the modification has potential influence on downstream or other related projects, this PR should be tested with those projects.
• CLA has been signed and all committers have signed the CLA in this PR.
  PHYBench-fullques_v1.json

[MaiziXiao]

LGTM