New lint: `too_many_lines_in_file` — warn when a source file exceeds a configurable line count

[tupe12334]

What it does

Warns when a source file contains more lines of code than a configurable threshold. Comment-only lines and blank lines would be excluded from the count (mirroring the behavior of too_many_lines for functions).

Advantage

• Encourages splitting large files into smaller, more focused modules
• Complements the existing too_many_lines lint, which operates at the function level, by also enforcing hygiene at the file/module level
• Large files are often a sign of poor separation of concerns and can hurt navigability and reviewability

Drawbacks

• This is inherently a style/restriction lint — reasonable threshold values vary widely by project
• Some files (e.g. generated code, large lookup tables) legitimately need to be long; users would need to suppress the lint for those
• Could be noisy on large existing codebases if adopted without tuning the threshold

Example

This lint fires on a file as a whole rather than a specific expression, so there is no before/after code rewrite. The diagnostic would look like:

warning: this file has too many lines (1200/1000)
  --> src/my_module.rs:1:1
   |
 1 | // src/my_module.rs
   | ^
   = help: consider splitting the module into smaller submodules

The intended fix is to split the file into smaller, more focused modules.

Comparison with existing lints

• too_many_lines — lints on individual functions exceeding a line threshold. This proposed lint operates at the file level, which is a coarser and complementary granularity.
• No existing Clippy lint checks overall file length.

Additional Context

• Should belong to the restriction group (opt-in), similar to too_many_lines
• The threshold should be configurable via clippy.toml, e.g. too-many-lines-in-file-threshold = 1000 (default could be 1000)
• Counting logic should match too_many_lines: skip blank lines and comment-only lines, count only lines with actual code