Do not propagate exception inside HTMLRewriter parser

[ryanking13]

When an exception is threw inside HTMLRewriter handler, that error marks the task as failure and make subsequent transforms fail. This PR changes it to catch the exception so it is not propagated other than the stream.

[ryanking13]

This comment is copied from "ReadableStream::serialize"

workerd/src/workerd/api/streams/readable.c++

Lines 857 to 862 in 754045d

	ioctx.addTask(
	ioctx.waitForDeferredProxy(pumpTo(js, kj::mv(sink), true)).catch_([](kj::Exception&& e) {
	// Errors in pumpTo() are automatically propagated to the source and destination. We don't
	// want to throw them from here since it'll cause an uncaught exception to be reported, even
	// if the application actually does handle it!
	}));

[codecov-commenter]

Codecov Report

❌ Patch coverage is 80.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 70.17%. Comparing base (a89ad60) to head (ccb8ecd).

Files with missing lines	Patch %	Lines
src/workerd/api/html-rewriter.c++	80.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #5976      +/-   ##
==========================================
- Coverage   70.17%   70.17%   -0.01%     
==========================================
  Files         407      407              
  Lines      107238   107241       +3     
  Branches    17985    17985              
==========================================
+ Hits        75253    75255       +2     
- Misses      21195    21196       +1     
  Partials    10790    10790              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[codspeed-hq]

CodSpeed Performance Report

Merging this PR will improve performance by 13.08%

_{Comparing gyeongjae/html-rewriter-posion-after-exception (ccb8ecd) with main (a89ad60)}

Summary

⚡ 1 improved benchmark
✅ 69 untouched benchmarks
⏩ 129 skipped benchmarks¹

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	simpleStringBody[Response]	21.6 µs	19.1 µs	+13.08%

Footnotes

1. 129 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[dom96]

Could this have any impact on existing code in prod? i.e. do we need to do this behind a new compat flag?

[jasnell]

@dom96 it should not have a negative impact. This is undoing part of another recent change that landed within the past two weeks.