feat(cli): per-user-turn token stats derived from the usage ledger (#379)

CHANGELOG.md

@@ -9,6 +9,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 - `tole doctor` (#378): tiered provider diagnostics — `offline` (wiring) and `catalog` (live `/models`, the model must be served); each blocker carries its fix command, the exit code is non-zero when the requested tier is not reached, and keys / URL credentials are never printed.
+- Per-user-turn token stats (#379): `tole status` lists input / output / cache-read / cache-write / reasoning tokens per turn and the status JSON (`GET /sessions/{id}/status`, `tole_session_status`) gains an additive `usage_report.turns` array. Derived from the durable usage ledger, so old sessions read the same way and nothing is stored twice; a value no step reported is `n/a` / `null`, never 0, and no elapsed time is invented.
 
 ### Changed
 - Memory loop injection hygiene (#384): the recalled batch is hard-capped at 24 items / 64 KiB and deduplicated; a malformed recall payload (non-array, or a hit without content) now fails the whole batch visibly instead of silently injecting fewer items; `tole serve`/MCP sessions report recall failures on stderr like the CLI does.
@@ -30,6 +31,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - `read_file` rejects an empty `path` with a clear message instead of a confusing "is a directory" error.
 
 ### Added
+- Per-user-turn token stats (#379): `tole status` lists input / output / cache-read / cache-write / reasoning tokens per turn and the status JSON (`GET /sessions/{id}/status`, `tole_session_status`) gains an additive `usage_report.turns` array. Derived from the durable usage ledger, so old sessions read the same way and nothing is stored twice; a value no step reported is `n/a` / `null`, never 0, and no elapsed time is invented.
 - `tole status` and the serve status JSON now report token usage, cache-hit rate and a per-step request-size split (system+tools vs history, stored per step under `tole_wire` in the usage record); `fact/mission` gains `cached_tokens`. Instrumentation only: the wire, budgets and durable entry shapes are unchanged (#211, part 1 of 3).
 
 ### Changed

README.md

@@ -298,7 +298,10 @@ the usage derived from the ledger for any session: prompt / completion /
 reasoning / cached tokens, the cache-hit rate, and the request size in
 characters split into system+tools vs history at the first and last step
 (#211; recorded per step under `tole_wire`; sessions recorded before it read
-`n/a`). Reported only — `--max-tokens` accounting is unchanged.
+`n/a`). It also lists per-user-turn stats (#379: input / output / cache-read /
+cache-write / reasoning tokens, `turns` in the status JSON), summed from the same
+ledger so they survive restarts; a value no step reported is `n/a` / `null`,
+never 0. Reported only — `--max-tokens` accounting is unchanged.
 
 ## Approval gates
 

crates/tole-cli/src/main.rs

@@ -2145,6 +2145,10 @@ fn status_command(sessions_dir: &Path, id: &str) -> Result<()> {
     for line in tole_cli::usage_report::usage_report(usage).render_lines() {
         println!("{line}");
     }
+    // Issue #379: per user-turn stats from the same ledger.
+    for line in tole_cli::usage_report::turn_lines(storage.entries(), usage) {
+        println!("{line}");
+    }
     Ok(())
 }
 

crates/tole-cli/src/serve.rs

@@ -675,7 +675,7 @@ fn route(state: &State, method: &str, path: &str, body: &str) -> (u16, serde_jso
                             Ok(guard) => (
                                 Some(guard.entries().len()),
                                 // Issue #211: additive usage report.
-                                tole_cli::usage_report::usage_report(guard.usages()).to_json(),
+                                tole_cli::usage_report::usage_json(guard.entries(), guard.usages()),
                             ),
                             // turn in flight — skip the count and report
                             Err(_) => (None, json!(null)),

crates/tole-cli/src/session_tools.rs

@@ -402,7 +402,7 @@ impl Tool for SessionStatusTool {
                 json!(guard.entries().len()),
                 // Issue #211: additive usage report (same object as the
                 // serve status JSON).
-                crate::usage_report::usage_report(guard.usages()).to_json(),
+                crate::usage_report::usage_json(guard.entries(), guard.usages()),
             ),
             Err(_) => (json!(null), json!(null)), // turn in flight
         };