Don't error when redownloading the same dandiset.yaml file

[jwodder]

Closes #819.

[codecov]

Codecov Report

Merging #824 (50079f7) into master (c28a4f0) will decrease coverage by 0.11%.
The diff coverage is 80.00%.

@@            Coverage Diff             @@
##           master     #824      +/-   ##
==========================================
- Coverage   85.31%   85.20%   -0.12%     
==========================================
  Files          61       61              
  Lines        6293     6298       +5     
==========================================
- Hits         5369     5366       -3     
- Misses        924      932       +8     

Flag	Coverage Δ
unittests	85.20% <80.00%> (-0.12%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
dandi/download.py	82.21% <80.00%> (-1.13%)	⬇️
dandi/dandiarchive.py	86.34% <0.00%> (-1.47%)	⬇️
dandi/cli/base.py	92.50% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c28a4f0...50079f7. Read the comment docs.

[yarikoptic]

this addresses that issue... but we have mess mixing all the yielding of records from _download_file and possibly raising exceptions, and then to have actual downloader to be a callback to pyout -- tried to detangle quickly but failed. All that code really needs RFing to become more consistent (no callbacks for pyout, just yielding records to pass updates to pyout, etc). But that would be a separate issue (well -- we have "good old" #249)