Centralize g3 status message formatting

Extract a new g3_status module in g3-cli that provides consistent formatting
for all 'g3:' prefixed system status messages.

Key changes:
- Add G3Status struct with methods for progress, done, failed, error, etc.
- Add Status enum with Done, Failed, Error, Resolved, Insufficient, NoChanges
- Add ThinResult struct in g3-core for semantic thinning data
- Update UiWriter trait with print_thin_result() method
- Refactor context thinning to return ThinResult instead of formatted strings
- Update all callers to use the new centralized formatting
- Session resume/decline messages now use G3Status
- Compaction status messages now use G3Status

This maintains clean separation of concerns: g3-core emits semantic data,
g3-cli handles all terminal formatting and colors.

crates/g3-core/tests/streaming_completion_test.rs

@@ -274,7 +274,7 @@ impl UiWriter for TrackingUiWriter {
     fn print_context_status(&self, _message: &str) {}
     fn print_g3_progress(&self, _message: &str) {}
     fn print_g3_status(&self, _message: &str, _status: &str) {}
-    fn print_context_thinning(&self, _message: &str) {}
+    fn print_thin_result(&self, _result: &g3_core::ThinResult) {}
 
     fn print_tool_header(&self, _tool_name: &str, _tool_args: Option<&serde_json::Value>) {
         // Count each tool call

crates/g3-core/tests/test_context_thinning.rs

@@ -69,14 +69,13 @@ fn test_thin_context_basic() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
-    // Should show the new format with percentage change
-    assert!(summary.contains("g3:"), "Summary was: {}", summary);
-    assert!(summary.contains("thinning context"));
-    assert!(summary.contains("[done]"));
+    // Should have made changes
+    assert!(result.had_changes, "Expected thinning to make changes");
+    assert!(result.chars_saved > 0, "Expected chars to be saved");
 
     // Check that the large tool results were replaced
     let first_third_end = context.conversation_history.len() / 3;
@@ -127,14 +126,13 @@ fn test_thin_write_file_tool_calls() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
-    // Should show the new format with percentage change
-    assert!(summary.contains("g3:"));
-    assert!(summary.contains("thinning context"));
-    assert!(summary.contains("[done]"));
+    // Should have made changes
+    assert!(result.had_changes, "Expected thinning to make changes");
+    assert!(result.chars_saved > 0, "Expected chars to be saved");
 
     // Check that the large content was replaced with a file reference
     let first_third_end = context.conversation_history.len() / 3;
@@ -189,14 +187,13 @@ fn test_thin_str_replace_tool_calls() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
-    // Should show the new format with percentage change
-    assert!(summary.contains("g3:"));
-    assert!(summary.contains("thinning context"));
-    assert!(summary.contains("[done]"));
+    // Should have made changes
+    assert!(result.had_changes, "Expected thinning to make changes");
+    assert!(result.chars_saved > 0, "Expected chars to be saved");
 
     // Check that the large diff was replaced with a file reference
     let first_third_end = context.conversation_history.len() / 3;
@@ -225,12 +222,12 @@ fn test_thin_context_no_large_results() {
     }
 
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    // Should report no large results found
-    assert!(summary.contains("g3:"));
-    assert!(summary.contains("thinning context"));
-    assert!(summary.contains("[no changes]"));
+    // Should report no changes (no large results found)
+    assert!(!result.had_changes, "Expected no changes");
+    assert_eq!(result.chars_saved, 0, "Expected no chars saved");
+    assert_eq!(result.leaned_count, 0, "Expected no messages thinned");
 }
 
 #[test]
@@ -256,11 +253,10 @@ fn test_thin_context_only_affects_first_third() {
     }
 
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    // Should show the new format with percentage change
-    assert!(summary.contains("g3:"));
-    assert!(summary.contains("[done]"));
+    // Should have made changes
+    assert!(result.had_changes, "Expected thinning to make changes");
 
     // Check that messages after the first third are NOT thinned
     let first_third_end = context.conversation_history.len() / 3;

crates/g3-core/tests/test_todo_context_thinning.rs

@@ -30,9 +30,9 @@ fn test_todo_read_results_not_thinned() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
     // Check that the TODO result was NOT thinned
     let first_third_end = context.conversation_history.len() / 3;
@@ -87,9 +87,9 @@ fn test_todo_write_results_not_thinned() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
     // Check that the TODO write result was NOT thinned
     let first_third_end = context.conversation_history.len() / 3;
@@ -135,14 +135,13 @@ fn test_non_todo_results_still_thinned() {
 
     // Trigger thinning at 50%
     context.used_tokens = 5000;
-    let (summary, _chars_saved) = context.thin_context(None);
+    let result = context.thin_context(None);
 
-    println!("Thinning summary: {}", summary);
+    println!("Thinning result: {:?}", result);
 
-    // Should show the new format with percentage change (indicating thinning happened)
-    assert!(summary.contains("g3:"), "Non-TODO results should be thinned");
-    assert!(summary.contains("thinning context"));
-    assert!(summary.contains("[done]"));
+    // Should have made changes (non-TODO results should be thinned)
+    assert!(result.had_changes, "Non-TODO results should be thinned");
+    assert!(result.chars_saved > 0, "Expected chars to be saved");
 
     // Check that the result was actually thinned
     let first_third_end = context.conversation_history.len() / 3;
@@ -184,7 +183,7 @@ fn test_todo_read_with_spaces_in_tool_name() {
 
     // Trigger thinning
     context.used_tokens = 5000;
-    let (_summary, _chars_saved) = context.thin_context(None);
+    let _result = context.thin_context(None);
 
     // Verify TODO result was not thinned
     let first_third_end = context.conversation_history.len() / 3;

crates/g3-core/tests/todo_staleness_test.rs

@@ -54,7 +54,7 @@ impl UiWriter for MockUiWriter {
     }
     fn print_g3_progress(&self, _message: &str) {}
     fn print_g3_status(&self, _message: &str, _status: &str) {}
-    fn print_context_thinning(&self, _message: &str) {}
+    fn print_thin_result(&self, _result: &g3_core::ThinResult) {}
     fn print_tool_header(&self, _tool_name: &str, _tool_args: Option<&serde_json::Value>) {}
     fn print_tool_arg(&self, _key: &str, _value: &str) {}
     fn print_tool_output_header(&self) {}