fix(scala): pass retry count to retryUntil delay function

[eric-zaharia]

🧭 What and Why

🎟 JIRA Ticket: API-422

RetryUntil.retryUntil passed the current delay value back into the delay function instead of the retry count. Because the seed is delay(0) = 0 and the delay functions are multiplicative (e.g. retries * 200), the result collapsed to 0 and fed itself 0 on every subsequent retry — so no backoff ever happened and polling hammered the API with zero delay.

The parameter was already named retries and every caller treats it as a retry count, so this was purely a wiring bug: the input's intended meaning ("retry count") never matched what was passed in ("current delay").

Changes included:

• Pass retryCount to delay(...) instead of the accumulated currentDelay.
• Drop the now-unused currentDelay accumulator and the delay(0) seed from the internal attempt helper. attempt is a local nested function, so this is not a public API change — the retryUntil signature and DEFAULT_DELAY: Long => Long are unchanged, and no call sites needed edits.

Resulting backoff now ramps as intended:

Delay function	Before	After
DEFAULT_DELAY (retries * 200, cap 5000)	0, 0, 0, …	0, 200, 400, …, 5000ms
pollEvent (retries * 1500, cap 5000)	0, 0, 0, …	0, 1500, 3000, 5000ms

🧪 Test

• yarn cli cts run scala passes. Note that CTS asserts polling correctness, not timing — the bug never broke correctness (it still polled, just with no delay), so CTS is a regression guard here rather than a direct validation of the backoff fix.

Found during review of #6312 (feat(scala): add withTransformation helpers).

[algolia-api-clients-automation-bot]

No code generated

If you believe code should've been generated, please, report the issue.

📊 Benchmark results

Benchmarks performed on the method using a mock server, the results might not reflect the real-world performance.

Language	Req/s
scala	23