feat(benchmark): add run_benchmark() with dataframe-first pipeline, metrics, and tests

[anshul23102]

Description

Adds a new run_benchmark() function as a database-free entry point for the
benchmark module. Unlike calc_benchmark(), this works without a BETYdb
connection, taking validated dataframes directly as input.

Files added:

• R/run_benchmark.R — dataframe-first pipeline with bm_validate(),
  align_by_time(), compute_metrics(), and plot_time_series()
• inst/testdata/sample_model.csv — sample model output
• inst/testdata/sample_obs.csv — sample observations
• tests/testthat/test-run_benchmark.R — unit tests for all four pipeline stages
• README.md — updated with quickstart example

Motivation and Context

The existing calc_benchmark() requires a full database connection which
makes it hard to test and use standalone. This adds a lightweight,
dataframe-first entry point — run_benchmark(model_df, obs_df) — for users
who want to quickly benchmark model output against observations without any
database setup. The pipeline follows a four-stage design: validate → align →
compute metrics → plot.

Review Time Estimate

• Immediately
• Within one week
• When possible

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• My change requires a change to the documentation.
• My name is in the list of CITATION.cff
• I agree that PEcAn Project may distribute my contribution under any or all of
  • the same license as the existing code,
  • and/or the BSD 3-clause license.
• I have updated the CHANGELOG.md.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• All new and existing tests passed.

[anshul23102]

Hi @dlebauer, just wanted to give you a quick update on PR #3892. All CI checks are now passing and I've refactored the API to be dataframe-first (run_benchmark(model_df, obs_df)) which directly aligns with the GSoC proposal I submitted.
The pipeline now has four clean stages: validate, align, compute metrics, and plot, each usable independently or through the orchestrator. Would love your feedback whenever you get a chance!