Add configurable retries with backoff to Prometheus Remote Write exporter

What problem do you want to solve?

Remote Write pushes are currently fail-fast: transient 429/5xx/network blips drop metric batches without retry or backpressure. That causes silent data loss during throttling or short outages.

Describe the solution you'd like

Add a configurable retry policy to the exporter: max attempts, exponential backoff with jitter, and a retryable status list (e.g., 429/408/5xx plus connection/timeouts). Log retry decisions and final failures. Defaults can be modest (e.g., 3 retries, small backoff) with max_retries=0 to disable.

Describe alternatives you've considered

Accept drops on error (current behavior): loses data during transient issues.
Rely on external queues/proxies: adds operational overhead and still needs exporter-side retries.
Push retry logic to remote endpoints: not always possible; exporters should be polite and back off themselves.

Additional Context

No response

Would you like to implement a fix?

Yes

Tip

_{React with 👍 to help prioritize this issue. Please use comments to provide useful context, avoiding +1 or me too, to help us triage it. Learn more here.}

Start by locating the Prometheus Remote Write exporter and the code path that sends metric batches. Review how failures are currently handled, then identify the relevant tests or add coverage for retry limits, backoff, retryable responses, logging, and disabled retries; done means transient failures follow the configured policy without dropping batches unexpectedly.