DEV: Update the ESLint caching mechanism

[pawelangelow]

What

The restore-keys is the fallback mechanism. When running on a branch like feature/RI-123/something:

• First looks for exact match: eslint-Linux-feature/RI-123/something cache
• If not found, falls back to any cache starting with eslint-Linux- (which includes eslint-Linux-main)

So what happens if we change .eslintrc.js? - That makes ESLint ignore the cache, lint all the files, and create a new cache.

The problem comes because the GitHub action doesn't update the cache if it has the same name (the action step returns: Post job cleanup. Cache hit occurred on the primary key eslint-Linux-feature-.... not saving cache.

There's a hack to do it manually - Go to Actions -> Caches and delete the old one, then after the first run, the cache will be OK. But this is manual work, so this PR solves this problem.

Another thing

For the sake of automation, now every merge in main would trigger the linter to prepare the cache. If that PR didn't update anything ESLint-related, it will reuse the cache, and it will be a very quick one.

Added workflow_dispatch as well, in case we want to run it manually.

---

Note

Improves ESLint workflow reliability and automation.

• Adds push on main and workflow_dispatch triggers to run lint automatically and on-demand
• Updates cache key/restore-keys to include hashFiles('.eslintrc.js') and github.ref_name, ensuring cache invalidates on ESLint config changes while retaining branch-specific reuse

^{Written by Cursor Bugbot for commit d179096. This will update automatically on new commits. Configure here.}