Fix NaN handling in formatPlanUnits

[pavankumar-vh]

Overview

Fix NaN handling in the formatPlanUnits function in common/src/util/freebuff-plan-summary.ts.

Bug Description

The function didn't validate that units is a finite number. If units was NaN or Infinity, Math.round(units * 10) / 10 would return NaN, causing the function to return 'NaN' as a string.

Fix

Added Number.isFinite() check to return '0' for invalid numbers.

Testing

No existing tests for this function, but the fix prevents incorrect behavior with invalid inputs.

Files Changed

• common/src/util/freebuff-plan-summary.ts - Added NaN validation

Scope

This change only touches common/ which is an approved contribution area per the Contributing Guide.

[codebuff-team]

Thanks for the small, focused fix. The change is correct in isolation: Math.round(NaN * 10) / 10 is NaN, and String(NaN) returns 'NaN', which would leak into UI-facing plan summaries. Guarding with Number.isFinite is a reasonable defensive addition, and it's scoped to common/, so it's in bounds for this mirror.

A couple of things a maintainer will likely want before porting:

1. Where does units actually become NaN/Infinity upstream? If you can point to a real call site (e.g. a division by zero or bad parse somewhere that feeds this function), that strengthens the case a lot - right now the PR is defensive-only, not tied to an observed bug.
2. Falling back to '0' silently could mask a real upstream data bug rather than surfacing it. Worth a quick gut-check on whether swallowing to '0' is the desired behavior versus, say, throwing or logging in dev.
3. No test was added. Since this file apparently has no existing tests, adding even one small test case (formatPlanUnits(NaN) === '0') would make this much easier to accept as-is rather than needing follow-up work.

Small, low-risk, and plausible - but light on justification and verification. Worth porting once the above is addressed or at least discussed.