Added a backfill in 7-day windows, and fetches now keep their data on a failure

Backfilling a long window (an early --since, or a long time offline)
hit the API's rate limit, which could still be exceeded after waiting
once, and then everything fetched so far was lost.

Before each Find a Tender source's worker starts, a start point more
than 7 days ago is now backfilled in 7-day windows bounded on both
sides, oldest first, each saved before the next is fetched. A window
that fails is saved as far as it got and fetched again from its start
after the source's interval. The worker then carries on from the end
of the last window.

A 429 is now waited out up to sources.<source>.rate_limit_retries
times (default 5) rather than once; fts-once --ignore-rate-limit uses
0. FetchAll returns what it fetched before an error along with it, and
the workers save it, so a failed run no longer loses its data.
findatender's FetchAll and FetchOnce take RequestOptions, which gains
Until (updatedTo), and RateLimitRetries replaces AllowRateLimit.

internal/fetcher/worker_findatender.go

@@ -17,7 +17,8 @@ import (
 const findATenderSource = "FindATender"
 
 // runFindATender polls Find a Tender every configured interval until ctx
-// is cancelled, passing what each run fetches to save.
+// is cancelled, passing what each run fetches to save, even when the run
+// fails partway. A long window is backfilled first (see backfill).
 func runFindATender(
 	ctx context.Context,
 	logger *slog.Logger,
@@ -29,29 +30,43 @@ func runFindATender(
 		return errors.New("fetcher: sources.find_a_tender.api_root is not set")
 	}
 
+	fetch := func(
+		ctx context.Context, since, until time.Time,
+	) ([]*types.Opportunity, error) {
+		return fetchFindATender(ctx, logger, conf, since, until)
+	}
+
+	since = backfill(
+		ctx,
+		logger,
+		findATenderSource,
+		conf.Interval,
+		fetch,
+		save,
+		since,
+	)
+
 	worker := util.Worker{
 		Logger:   tagged(logger),
 		Name:     findATenderSource,
 		Interval: conf.Interval,
 		Task: func(ctx context.Context, since time.Time) error {
-			opportunities, err := fetchFindATender(ctx, logger, conf, since)
-			if err != nil {
-				return err
-			}
-			return save(ctx, logger, opportunities)
+			opportunities, err := fetch(ctx, since, time.Time{})
+			return saveFetched(ctx, logger, save, opportunities, err)
 		},
 	}
 	return worker.Run(ctx, since)
 }
 
-// fetchFindATender fetches everything updated since since from Find a
-// Tender. Finding nothing returns ErrNoNewReleases, so the worker still moves
-// its checkpoint on.
+// fetchFindATender fetches everything updated from since up to until (zero:
+// no bound) from Find a Tender. If it fails partway, what it fetched before
+// the error is returned along with it. Finding nothing returns
+// ErrNoNewReleases, so the worker still moves its checkpoint on.
 func fetchFindATender(
 	ctx context.Context,
 	logger *slog.Logger,
 	conf *config.JSCSourceFindATender,
-	since time.Time,
+	since, until time.Time,
 ) ([]*types.Opportunity, error) {
 	opportunities, err := util.AwaitTask(
 		ctx,
@@ -60,17 +75,18 @@ func fetchFindATender(
 				conf.APIRoot, findatender.WithLogger(logger),
 			)
 			defer findATender.Shutdown()
-			return findATender.With(ctx).FetchAll(since, true)
+			return findATender.With(ctx).FetchAll(findatender.RequestOptions{
+				Since:            since,
+				Until:            until,
+				RateLimitRetries: conf.RateLimitRetries,
+			})
 		},
 	)
 	if errors.Is(err, findatender.ErrNoReleases) {
 		return nil, ErrNoNewReleases
 	}
-	if err != nil {
-		return nil, err
-	}
 
 	tagged(logger).Info("fetched opportunities", "count", len(opportunities))
 
-	return opportunities, nil
+	return opportunities, err
 }