perf(linter/plugins): optimize getTokens() and other methods

[lilnasy]

• Removes branching in some cases where it was used to calculate slices.
• Avoids indexing into arrays past their length.
• Renames nodeTokens to tokenList. In the first few methods implemented, they contained tokens of the given node. Not the case for later methods, but the misnomer stayed.
• Improves consistency in comments. Clarifies that binary searches yield tokens before or after a certain point, they are not necessarily within the range we care about.

[Copilot]

Pull request overview

This PR optimizes token retrieval methods in the linter plugin by improving performance through several targeted changes: removing unnecessary branching, preventing out-of-bounds array access, renaming variables for clarity, and improving comment consistency.

Key changes:

• Optimizes getTokens() to avoid unnecessary array operations when filtering with count limits and adds bounds checking to prevent indexing past array length
• Refactors binary search in getLastToken() with a cleaner algorithm and adds explicit bounds checking
• Renames nodeTokens variable to tokenList throughout the file for better semantic accuracy

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
apps/oxlint/test/tokens.test.ts	Fixes incorrect range values in tests (corrects [0, 11] to [0, 10] to match actual string length)
apps/oxlint/src-js/plugins/tokens.ts	Refactors token retrieval methods with performance optimizations: adds bounds checking, removes branching in favor of nullish coalescing, optimizes filtering loops, and renames nodeTokens to tokenList for clarity

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.