Fix lag metric for grouper

[samarth-gupta-traceable]

Description

I see 2 potential issues with current arrival lag reporting:

1. original objective on 'arrival.lag' was to report lag at start of job. overtime this has been changed it seems, and currently it reports lag when trace is about to be emitted.
   hence we are currently reporting lag including the time span spent in rawspangrouper.

2. another issue is that, we are determining the lag based on trace start time which is set of grouper as currenttime. which does not give the correct value of lag. as lag represents how much delay as span has faced till now in the data pipeline.

this PR it trying to address only the 2nd point. as of now.

For now i am of the opinion we can consider the metrics been reported as just lag instead of arrival lag. Grouper is already reporting at end of job. I have made changes to move enricher reporting also to end of enrichment.
any thoughts @ravisingal @surajpuvvada ?

[codecov]

Codecov Report

Merging #283 (5c8a554) into main (30f0a7b) will decrease coverage by 0.01%.
The diff coverage is 75.00%.

❗ Current head 5c8a554 differs from pull request most recent head 56ea3d7. Consider uploading reports for the commit 56ea3d7 to get more accurate results

@@             Coverage Diff              @@
##               main     #283      +/-   ##
============================================
- Coverage     79.07%   79.06%   -0.02%     
  Complexity     1231     1231              
============================================
  Files           110      110              
  Lines          4861     4863       +2     
  Branches        440      441       +1     
============================================
+ Hits           3844     3845       +1     
  Misses          813      813              
- Partials        204      205       +1     

Flag	Coverage Δ
unit	79.06% <75.00%> (-0.02%)	⬇️

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

Impacted Files	Coverage Δ
...race/core/rawspansgrouper/TraceEmitPunctuator.java	74.13% <75.00%> (-0.43%)	⬇️

---

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 30f0a7b...56ea3d7. Read the comment docs.

[samarth-gupta-traceable]

@rish691 @kotharironak I am not exactly sure what does getReceivedTimeMillis on a span represent. does it tell us when was span received at span normalizer or platform agent or tracing agent ?

[github-actions]

Unit Test Results

  73 files  ±0    73 suites  ±0   1m 4s ⏱️ +3s
390 tests ±0  390 ✔️ ±0  0 💤 ±0  0 ❌ ±0 

Results for commit 56ea3d7. ± Comparison against base commit 30f0a7b.

[samarth-gupta-traceable]

@rish691 @kotharironak I am not exactly sure what does getReceivedTimeMillis on a span represent. does it tell us when was span received at span normalizer or platform agent or tracing agent ?

looks like it is time when span was received at normalizer. https://github.com/hypertrace/hypertrace-ingester/blob/main/span-normalizer/span-normalizer/src/main/java/org/hypertrace/core/spannormalizer/jaeger/JaegerSpanNormalizer.java#L116