[Fix] Resolve CodeFactor lint findings

[antoniojbt]

Motivation

• A CodeFactor/lintr run on the PR reported multiple lint issues (indentation, object name length/style and mis-parsed shell wrapper) which caused the check to fail.
• The intent is to fix lint violations without breaking public APIs or changing runtime behaviour of exported functions.

Description

• Fixed indentation and formatting issues in test files under tests/testthat/ to satisfy the configured 2-space/hanging-indent rules.
• Updated R/epi_stats_numeric.R to use an internal snake_case variable n_non_na while preserving the public output column name n_nonNA with a targeted # nolint: object_name_linter suppression.
• Shortened private/helper names used only in tests (e.g. expected_project_template_entries -> template_entries, expect_project_template_entries -> expect_template_entries) to address object_length_linter warnings.
• Excluded the bash R wrapper scripts/rscript_env_caller.R from R linters by adding it to .lintr exclusions, and added a couple of # nolint annotations in R/eda_synthetic.R to silence deliberate style exceptions.

Testing

• Inspected the CodeFactor annotations via the GitHub API and reproduced the lint list; used that output to target fixes.
• Ran bash -n scripts/rscript_env_caller.R to validate shell syntax which succeeded.
• Ran git diff --check to ensure no whitespace/merge conflict markers remain which succeeded.
• Attempted to run scripts/rscript_env_caller.R -e "lintr::lint_package()" but it could not complete because the repo-configured Rscript binary (/root/apps/miniforge3/envs/episcout/bin/Rscript) is not available in this environment; lintr checks are the reason for the changes and should pass in CI where the R environment exists.

---

Codex Task