TAP5-2837: Improve JSON serialization performance

[benweidig]

Overview

JSONObject/JSONArray's toString() and toCompactString() are slower than necessary due to synchronization overhead and per-String allocation in the print path.

My sandbox implementation at https://github.com/benweidig/tapestry-json-improvements (non-breaking, verified byte-for-byte against current output) shows a ~4-6x speedup across payload sizes and both compact/pretty formatting, with no public API changes.

Changes

• Deprecated JSONStringer and inlined its escaping loop directly into JSONObject. There is some code duplication, but it keeps the call sites monomorphic and should allow for better C2 optimization in the JVM.
• Hex escaping is done with manual nibble extraction pls a lookup-table instead of String.format calls
• Droped CharArrayWriter/PrintWriter and build directlty into StringBuilder
• Pre-size output buffers

Results

Graphs: JMH Visalizer

End-to-end cost: JSONSerializationBenchmark (toString() / toCompactString())

Payload size	compact: baseline (µs/op)	compact: improved (µs/op)	speedup	pretty: baseline (µs/op)	pretty: improved (µs/op)	speedup
10	15.10 ± 0.34	3.57 ± 0.27	4.2×	33.30 ± 1.03	6.16 ± 0.07	5.4×
100	186.83 ± 22.62	33.14 ± 0.52	5.6×	349.52 ± 4.62	57.95 ± 0.38	6.0×
1000	1777.78 ± 34.98	388.31 ± 4.70	4.6×	3469.37 ± 120.15	629.63 ± 17.88	5.5×

Isolated escaping cost: StringQuotingBenchmark (JSONObject.quote(String))

Profile	Baseline (ns/op)	Improved (ns/op)	Speedup
PLAIN_ASCII	175.74	81.08	2.2×
QUOTES_AND_SLASHES	168.55	82.93	2.0×
CONTROL_CHARS	182.26	80.45	2.3×
UNICODE_HEAVY	611.93	105.65	5.8×