add pyserini onnx encoder support

[ArthurChen189]

Added support for Pyserini usage of ONNX on-the-fly query encoder
Related PR

[codecov-commenter]

Codecov Report

Patch coverage: 2.22% and project coverage change: -0.13 ⚠️

Comparison is base (4aeb3ef) 59.17% compared to head (a015837) 59.05%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #2113      +/-   ##
============================================
- Coverage     59.17%   59.05%   -0.13%     
  Complexity     1200     1200              
============================================
  Files           194      194              
  Lines         11352    11377      +25     
  Branches       1488     1488              
============================================
+ Hits           6718     6719       +1     
- Misses         4146     4170      +24     
  Partials        488      488              

Impacted Files	Coverage Δ
...nserini/analysis/HuggingFaceTokenizerAnalyzer.java	77.77% <ø> (ø)
...main/java/io/anserini/eval/RelevanceJudgments.java	56.94% <ø> (ø)
...c/main/java/io/anserini/search/SimpleSearcher.java	69.66% <ø> (ø)
...in/java/io/anserini/search/query/QueryEncoder.java	0.00% <ø> (ø)
...uery/SpladePlusPlusEnsembleDistilQueryEncoder.java	0.00% <0.00%> (ø)
...ch/query/SpladePlusPlusSelfDistilQueryEncoder.java	0.00% <0.00%> (ø)
.../io/anserini/search/query/UniCoilQueryEncoder.java	0.00% <0.00%> (ø)
...va/io/anserini/search/topicreader/TopicReader.java	66.29% <ø> (ø)
.../java/io/anserini/search/SimpleImpactSearcher.java	71.42% <8.33%> (-7.30%)	⬇️

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[lintool]

Can we have some test cases? At the very least, let's not have coverage go down?

[lintool]

lgtm

[lintool]

hey @ArthurChen189 I just got this issue:

Results :

Failed tests:   testOnnxEncoder(io.anserini.search.SimpleImpactSearcherTest): expected:<3.053449> but was:<3.05345>

Tests run: 675, Failures: 1, Errors: 0, Skipped: 0

This is on my M1 Mac, so floating point tolerance issues... should we adjust precision?