[Logs UI] Exponential date range adjustments can go out of range, cause a loop of 500 errors

[jasonrhodes]

In the stream view, when you have a date range selected and you reach the edges, we provide some buttons to "extend time frame" which somewhat exponentially increases the range which it expands by. If you're filtering for a value in the search box at the same time, it's easy to click that button many times in a row as no new entries are loaded, which eventually causes an out of range query error.

Easiest way to reproduce:

1. Choose any log line, find it's _id and click the filter by button to filter by only that _id value (you should have only one entry
2. Choose an "Absolute time" for the "From" part of the time range
3. Click the "extend time frame" at the top to go back in time, watch as it increases quite rapidly with each click (my example did 7 years, 21 years, 44 years, then started failing).

Thanks to @sqren for the find and report!

{
  "statusCode": 500,
  "error": "Internal Server Error",
  "message": "[parse_exception] failed to parse date field [-2225168973352] with format [epoch_millis]: [failed to parse date field [-2225168973352] with format [epoch_millis]] (and) [parse_exception] failed to parse date field [-2225168973352] with format [epoch_millis]: [failed to parse date field [-2225168973352] with format [epoch_millis]] (and) [parse_exception] failed to parse date field [-2225168973352] with format [epoch_millis]: [failed to parse date field [-2225168973352] with format [epoch_millis]] (and) [parse_exception] failed to parse date field [-2225168973352] with format [epoch_millis]: [failed to parse date field [-2225168973352] with format [epoch_millis]]"
}

[elasticmachine]

Pinging @elastic/logs-metrics-ui (Team:logs-metrics-ui)