fix(ingest): use lossy collections

[anshbansal]

Checklist

• The PR conforms to DataHub's Contributing Guideline (particularly Commit Message Format)
• Links to related issues (if applicable)
• Tests for the changes have been added/updated (if applicable)
• Docs related to the changes have been added/updated (if applicable). If a new feature has been added a Usage Guide has been added for the same.
• For any breaking change/potential downtime/deprecation/big changes an entry has been made in Updating DataHub

[codecov]

Codecov Report

Attention: Patch coverage is 96.29630% with 2 lines in your changes missing coverage. Please review.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
...gestion/src/datahub/ingestion/source/abs/report.py	0.00%	2 Missing ⚠️

Files with missing lines	Coverage Δ
...ngestion/source/aws/sagemaker_processors/common.py	97.77% <100.00%> (+0.05%)	⬆️
...ub/ingestion/source/bigquery_v2/bigquery_report.py	100.00% <100.00%> (ø)
.../src/datahub/ingestion/source/delta_lake/report.py	83.33% <100.00%> (ø)
.../src/datahub/ingestion/source/dynamodb/dynamodb.py	80.97% <100.00%> (+0.09%)	⬆️
...ion/src/datahub/ingestion/source/elastic_search.py	56.27% <100.00%> (+0.19%)	⬆️
...datahub/ingestion/source/iceberg/iceberg_common.py	97.43% <100.00%> (+0.02%)	⬆️
...tion/src/datahub/ingestion/source/identity/okta.py	80.43% <100.00%> (+0.07%)	⬆️
...estion/src/datahub/ingestion/source/kafka/kafka.py	94.05% <100.00%> (+0.02%)	⬆️
...c/datahub/ingestion/source/kafka_connect/common.py	99.02% <100.00%> (+<0.01%)	⬆️
...ata-ingestion/src/datahub/ingestion/source/ldap.py	90.78% <100.00%> (+0.04%)	⬆️
... and 11 more

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6cb5191...ed43eb8. Read the comment docs.

[hsheth2]

Actually the TopKDict is really only usable for value types where sorting is a reasonable thing to do. we're using it in a couple spots with string mem_size, which probably won't work properly

[anshbansal]

Need to update the integration tests

[anshbansal]

kafka connect failure is being handled in another PR