correct place to initialize mrc_index_to_particles array

[vineetbansal]

Moving the mrc_index_to_particles logic to after the size filtering/row truncation seems to me to be the right thing to do (and one that involves the least disruption), and fixes the problem of passing out-of-range indices to ImageSource.set_metadata method (which its currently tolerant of, and ignores).

Before this change (for a 2-file, 400 particle source, out of which 100 are filtered out on size criteria):

self.mrc_index_to_particles:
0 = [0, 1, 2, ... 199]
1 = [200, 201, ... 399]

After this change:

self.mrc_index_to_particles:
0 = [0, 1, 2, ... 146]
1 = [147, 148, ... 299]

[codecov]

Codecov Report

Merging #906 (5df2baf) into develop (ccd044c) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff            @@
##           develop     #906   +/-   ##
========================================
  Coverage    88.17%   88.18%           
========================================
  Files          119      119           
  Lines        10118    10118           
========================================
+ Hits          8922     8923    +1     
+ Misses        1196     1195    -1     

Impacted Files	Coverage Δ
src/aspire/source/coordinates.py	92.20% <100.00%> (ø)

... and 1 file with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[garrettwrong]

Looks good to me, thanks!