chore(release): v0.7.0 — version bump, changelog, docs sync

CHANGELOG.md

@@ -4,7 +4,7 @@ All notable changes to Ripple are documented here. The format is based on
 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project
 adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [Unreleased]
+## [0.7.0] - 2026-08-18
 
 ### Added
 

README.md

@@ -100,7 +100,7 @@ Everything below is real output against a shipped example project in `tests/fixt
 ```bash
 npm install -g @alimaandev/ripple
 
-ripple version                  # → 0.6.0
+ripple version                  # → 0.7.0
 ```
 
 Or install from source:
@@ -132,7 +132,7 @@ ripple analyze src/authentication/login.ts
 Expected output:
 
 ```text
-ripple · impact analysis · v0.6.0
+ripple · impact analysis · v0.7.0
 
 File        src/authentication/login.ts
 Risk        MEDIUM · 34.7/100 ███░░░░░░░
@@ -165,7 +165,7 @@ ripple graph
 ```
 
 ```text
-ripple · dependency graph · v0.6.0
+ripple · dependency graph · v0.7.0
 
 Files            25
 Edges            28

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@alimaandev/ripple",
-  "version": "0.6.0",
+  "version": "0.7.0",
   "description": "Dependency impact analysis for TypeScript and JavaScript. Know the blast radius of any change — before you commit.",
   "keywords": [
     "cli",

src/cache/parsed.ts

@@ -1,3 +1,3 @@
 import { createHash } from "node:crypto";
 import fs from "node:fs/promises";
 import path from "node:path";
@@ -13,13 +13,13 @@
  * Parsing a file with ts-morph (imports, exports, symbols) is the dominant
  * cost of an analysis run. For every discovered file we reuse the previously
  * parsed surface when the file is provably unchanged. Freshness is checked
- * with a cheap mtime+size match first — the steady-state hot path touches
- * neither the file contents nor the hasher — and the cache is only rewritten
+ * with a cheap mtime+size match first — the steady-state hot path touches
+ * neither the file contents nor the hasher — and the cache is only rewritten
  * to disk when something actually changed, so repeated runs are near-free.
  *
  * The cache is keyed by the discovery-affecting config (include/ignore), so
  * changing discovery invalidates it wholesale. The parsed surface is only a
- * function of file content — resolution, cycles and risk are recomputed
+ * function of file content — resolution, cycles and risk are recomputed
  * fresh on every run, so cached runs are byte-identical to cold runs.
  *
  * The cache lives in `<rootDir>/.ripple/cache/` and is written atomically.
@@ -79,7 +79,7 @@
   return path.join(rootDir, CACHE_FILE);
 }
 
-/** Load the cache manifest; any problem (missing, corrupt, stale) → empty. */
+/** Load the cache manifest; any problem (missing, corrupt, stale) → empty. */
 export async function loadParsedCache(
   rootDir: string,
   configHash: string,
@@ -139,7 +139,7 @@
  * whose content changed since the last run. The result is ordered exactly
  * like `filePaths`, so downstream output is identical to a cold run.
  *
- * When nothing changed, the on-disk cache is not rewritten at all — the
+ * When nothing changed, the on-disk cache is not rewritten at all — the
  * steady-state hot path is stat + JSON parse + graph rebuild.
  */
 export async function loadParsedFiles(options: {