Use requests stream and shutil.copyfileobj to constrain memory usage during resource copy

[whargrove]

Link to Relevant Issue

This pull request resolves #235

Description of Changes

• When copying resources from a remote origin over HTTP(S) prefer to stream the response body and copy chunks into the destination file insead of loading the entire file into memory first before writing.
• Before the change, running resource_copy("uri") in python REPL would use unbounded rss (up to file size). After the change, running the same in python REPL uses ~300M rss.

[codecov[bot]]

Codecov Report

Merging #236 (0964eb4) into main (6c1c1ff) will increase coverage by 0.45%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main     #236      +/-   ##
==========================================
+ Coverage   71.67%   72.12%   +0.45%     
==========================================
  Files          50       50              
  Lines        3329     3376      +47     
==========================================
+ Hits         2386     2435      +49     
+ Misses        943      941       -2     

Impacted Files	Coverage Δ
cdp_backend/tests/utils/test_file_utils.py	100.00% <100.00%> (ø)
cdp_backend/utils/file_utils.py	90.47% <100.00%> (+2.18%)	:arrow_up: