Show a pretty markdown table in formatter ecosystem checks (#6496)

**Summary** The formatter ecosystem checks will now print a markdown table you can copy&paste into your PR description. 

![image](https://github.com/astral-sh/ruff/assets/6826232/80289ed9-9d2b-400e-a994-de63dca0b065)

copied markdown:

| project      | similarity index |
|--------------|------------------|
| build        | 0.75623          |
| cpython      | 0.75989          |
| django       | 0.99784          |
| transformers | 0.99470          |
| typeshed     | 0.74853          |
| warehouse    | 0.99585          |
| zulip        | 0.99702          |

raw markdown:
```markdown
| project      | similarity index |
|--------------|------------------|
| build        | 0.75623          |
| cpython      | 0.75989          |
| django       | 0.99784          |
| transformers | 0.99470          |
| typeshed     | 0.74853          |
| warehouse    | 0.99585          |
| zulip        | 0.99702          |
```

.github/workflows/ci.yaml

@@ -338,6 +338,6 @@ jobs:
       - name: "Formatter progress"
         run: scripts/formatter_ecosystem_checks.sh
       - name: "Github step summary"
-        run: grep "similarity index" target/progress_projects_log.txt | sort > $GITHUB_STEP_SUMMARY
+        run: cat target/progress_projects_stats.txt > $GITHUB_STEP_SUMMARY
       - name: "Remove checkouts from cache"
         run: rm -r target/progress_projects

crates/ruff_dev/src/format_dev.rs

@@ -22,7 +22,7 @@ use std::panic::catch_unwind;
 use std::path::{Path, PathBuf};
 use std::process::ExitCode;
 use std::time::{Duration, Instant};
-use std::{fmt, fs, io};
+use std::{fmt, fs, io, iter};
 use tempfile::NamedTempFile;
 use tracing::{debug, error, info, info_span};
 use tracing_indicatif::span_ext::IndicatifSpanExt;
@@ -187,6 +187,9 @@ pub(crate) struct Args {
     /// Write all log messages (same as cli) to this file
     #[arg(long)]
     pub(crate) log_file: Option<PathBuf>,
+    /// Write a markdown table with the similarity indices to this file
+    #[arg(long)]
+    pub(crate) stats_file: Option<PathBuf>,
     /// Assert that there are exactly this many input files with errors. This catches regressions
     /// (or improvements) in the parser.
     #[arg(long)]
@@ -302,6 +305,8 @@ fn format_dev_multi_project(args: &Args) -> anyhow::Result<bool> {
         None => None,
     };
 
+    let mut results = Vec::new();
+
     for project_path in project_paths {
         debug!(parent: None, "Starting {}", project_path.display());
 
@@ -332,6 +337,7 @@ fn format_dev_multi_project(args: &Args) -> anyhow::Result<bool> {
                     write!(error_file, "{}", result.display(args.format)).unwrap();
                     error_file.flush().unwrap();
                 }
+                results.push(result);
 
                 pb_span.pb_inc(1);
             }
@@ -353,6 +359,35 @@ fn format_dev_multi_project(args: &Args) -> anyhow::Result<bool> {
         duration.as_secs_f32(),
     );
 
+    if let Some(stats_file) = &args.stats_file {
+        results.sort_by(|result1, result2| result1.name.cmp(&result2.name));
+        let project_col_len = results
+            .iter()
+            .map(|result| result.name.len())
+            .chain(iter::once("project".len()))
+            .max()
+            .unwrap_or_default();
+        let mut stats_file = BufWriter::new(File::create(stats_file)?);
+        writeln!(
+            stats_file,
+            "| {:<project_col_len$} | similarity index |",
+            "project"
+        )?;
+        writeln!(
+            stats_file,
+            "|-{:-<project_col_len$}-|------------------|",
+            ""
+        )?;
+        for result in results {
+            writeln!(
+                stats_file,
+                "| {:<project_col_len$} | {:.5}          |",
+                result.name,
+                result.statistics.similarity_index()
+            )?;
+        }
+    }
+
     if let Some(files_with_errors) = args.files_with_errors {
         if total_syntax_error_in_input != files_with_errors {
             error!(
@@ -433,7 +468,13 @@ fn format_dev_project(
 
     let duration = start.elapsed();
 
+    let name = files[0]
+        .file_name()
+        .unwrap_or(files[0].as_os_str())
+        .to_string_lossy()
+        .to_string();
     Ok(CheckRepoResult {
+        name,
         duration,
         file_count: formatted_counter,
         diagnostics,
@@ -511,6 +552,7 @@ fn diff_show_only_changes(
 }
 
 struct CheckRepoResult {
+    name: String,
     duration: Duration,
     file_count: usize,
     diagnostics: Vec<Diagnostic>,

scripts/formatter_ecosystem_checks.sh

@@ -55,10 +55,11 @@ git -C "$dir/cpython" checkout 45de31db9cc9be945702f3a7ca35bbb9f98476af
 # Uncomment if you want to update the hashes
 # for i in "$dir"/*/; do git -C "$i" switch main && git -C "$i" pull && echo "# $(basename "$i") $(git -C "$i" rev-parse HEAD)"; done
 
-time cargo run --bin ruff_dev -- format-dev --stability-check --error-file "$target/progress_projects_errors.txt" \
-  --log-file "$target/progress_projects_log.txt" --files-with-errors 25 --multi-project "$dir" || (
+time cargo run --bin ruff_dev -- format-dev --stability-check \
+  --error-file "$target/progress_projects_errors.txt" --log-file "$target/progress_projects_log.txt" --stats-file "$target/progress_projects_stats.txt" \
+  --files-with-errors 25 --multi-project "$dir" || (
   echo "Ecosystem check failed"
   cat "$target/progress_projects_log.txt"
   exit 1
 )
-grep "similarity index" "$target/progress_projects_log.txt" | sort
+cat "$target/progress_projects_stats.txt"