feat: update WASM runtime probe

[antsukanova]

Description

Replace the WASM-to-JS timing ratio with a WASM-only runtime benchmark.

The previous probe could report slow WASM when JavaScript was slow but WASM was healthy. The new worker measures WASM add, divide, and square-root operations and compares their timings to identify fast and slow WASM runtimes.

This change:

• Runs the benchmark in a Web Worker
• Uses five trials and median timings
• Returns OK, SLOW, UNCERTAIN, UNKNOWN, or DISABLED
• Returns one reason for unknown or uncertain results
• Groups useful numeric data under measurements
• Distinguishes worker startup, timeout, runtime, benchmark, and measurement failures
• Improves naming and documentation around benchmark calculations
• Updates unit coverage for probe

Related work:

• SPARK-844160

This change implements...

• A new feature
• A bug fix
• Other (please specify):

Is this a breaking change?

• Yes
• No

Test evidence

• yarn test passed
• Build, ESLint, Prettier, and spelling checks passed
• Real browser-worker validation part: https://confluence-eng-gpk2.cisco.com/conf/spaces/webexmedia/pages/881617568/WASM+probe+metrics+and+status

I certify that...

• All relevant unit and integration tests have passed and/or have been updated according to this change.

Generative AI (GAI) Usage Disclosure

• Cisco approved GAI tool/IDE was used for coding

• No GAI

• Manual Draft with GAI Refinement

• GAI Draft with Manual Customization

• GAI Generated Code

Additional GAI Usage Details:

Cursor was used to review naming, simplify API documentation, refactor result handling, and extend unit-test coverage.