[stdlib] Change Slice, CollectionOfOne and ReversedCollection to use _read/_modify

[airspeedswift]

Another swathe.

[airspeedswift]

@swift-ci please benchmark

[swift-ci]

Build comment file:

Performance: -O

TEST	OLD	NEW	DELTA	RATIO
Regression
StringMatch	7953	19411	+144.1%	0.41x
RomanNumbers	67041	111578	+66.4%	0.60x
CSVParsing2	2275	2869	+26.1%	0.79x
Improvement
Breadcrumbs.CopyUTF16CodeUnits.ASCII	48	43	-10.4%	1.12x
ArrayAppendAsciiSubstring	14112	12862	-8.9%	1.10x
StringEqualPointerComparison	589	538	-8.7%	1.09x
Breadcrumbs.UTF16ToIdxRange.longASCII	81	75	-7.4%	1.08x (?)

Code size: -O

TEST	OLD	NEW	DELTA	RATIO
Regression
CSVParsing.o	31465	33945	+7.9%	0.93x
ArrayAppend.o	39272	40000	+1.9%	0.98x
StringEdits.o	12935	13135	+1.5%	0.98x
Substring.o	18215	18415	+1.1%	0.99x

Performance: -Osize

TEST	OLD	NEW	DELTA	RATIO
Regression
StringMatch	8437	20208	+139.5%	0.42x
RomanNumbers	69736	117093	+67.9%	0.60x
CSVParsing2	2592	2969	+14.5%	0.87x
Improvement
Breadcrumbs.CopyUTF16CodeUnits.ASCII	48	43	-10.4%	1.12x
PrefixAnySeqCRangeIterLazy	158	142	-10.1%	1.11x
StringEqualPointerComparison	564	512	-9.2%	1.10x
Breadcrumbs.UTF16ToIdxRange.longASCII	82	76	-7.3%	1.08x (?)

Code size: -Osize

TEST	OLD	NEW	DELTA	RATIO
Regression
CSVParsing.o	31905	34009	+6.6%	0.94x

Performance: -Onone

TEST	OLD	NEW	DELTA	RATIO
Regression
CSVParsing2	4538	7293	+60.7%	0.62x
SuffixArrayLazy	35074	47724	+36.1%	0.73x
SuffixCountableRangeLazy	36975	49872	+34.9%	0.74x
StringMatch	48902	62475	+27.8%	0.78x
RomanNumbers	2194284	2707185	+23.4%	0.81x
DropFirstArrayLazy	109499	134158	+22.5%	0.82x
PrefixArrayLazy	107739	130388	+21.0%	0.83x
DropFirstCountableRangeLazy	119277	140325	+17.6%	0.85x
PrefixCountableRangeLazy	119562	139889	+17.0%	0.85x
DropLastCountableRangeLazy	41327	47064	+13.9%	0.88x
PrefixAnyCollectionLazy	183039	207679	+13.5%	0.88x
DropLastArrayLazy	38347	42584	+11.0%	0.90x
Improvement
ArrayAppendLatin1Substring	46222	36394	-21.3%	1.27x
ArrayAppendAsciiSubstring	46170	36509	-20.9%	1.26x
ArrayAppendUTF16Substring	45965	36567	-20.4%	1.26x

How to read the data

The tables contain differences in performance which are larger than 8% and differences in code size which are larger than 1%.

If you see any unexpected regressions, you should consider fixing the
regressions before you merge the PR.

Noise: Sometimes the performance results (not code size!) contain false
alarms. Unexpected regressions which are marked with '(?)' are probably noise.
If you see regressions which you cannot explain you can try to run the
benchmarks again. If regressions still show up, please consult with the
performance team (@eeckstein).

Hardware Overview

  Model Name: Mac Pro
  Model Identifier: MacPro6,1
  Processor Name: 8-Core Intel Xeon E5
  Processor Speed: 3 GHz
  Number of Processors: 1
  Total Number of Cores: 8
  L2 Cache (per Core): 256 KB
  L3 Cache: 25 MB
  Memory: 64 GB

--------------