RUMM-610 Hotfix for memory leaks on data upload

[ncreated]

What and why?

🐛 This hotfix is addressing #178 - where constantly growing memory usage is reported due to intensive logging.

With intensive logging comes frequent upload... and what I noticed is that every upload results with a memory leak. Here, allocations graph for 10 uploads (each batch was ~1MB):

How?

The reason of this leak, was an internal caching applied by the URLSession - by default, the session's configuration.urlCache is not nil, which results with some representation of the URLRequest being kept in memory (together with its .httpBody):

This started being visible after #65, where the ?batch_time= query parameter was introduced and cache keys become unique for every new request.

The fix is simple - we need to opt-out from using the .urlCache for the URLSession. After this, the allocation graph is no longer growing infinitely (same 10 batches, ~1MB each - no more "Live" blobs):

Review checklist

• Feature or bugfix MUST have appropriate tests (unit, integration)
• Make sure each commit and the PR mention the Issue number or JIRA reference

[mariusc83]

Was thinking about side effects ...but as we are only posting data I guess this cannot harm us.