SILOptimizer: some compile time fixes

[eeckstein]

Disable some expensive optimizations for huge functions in SimplifyCFG and the performance inliner.
Those optimizations are not strictly linear with the number of blocks and cause compile time issues if the function is really huge.

SR-10209
rdar://problem/49522869

[eeckstein]

@swift-ci smoke test

[eeckstein]

@swift-ci benchmark

[atrick]

For the record, I'm all for disabling the SPA heuristic for large functions (I can't understand how the heuristic works anyway). But optimizations should be written such that they can be applied to any function size and either scale linearly or halt the analysis at a certain point rather than disabling an optimization completely based on an arbitrary function size.

[swift-ci]

Performance: -O

TEST	OLD	NEW	DELTA	RATIO
Regression
DataAppendDataLargeToLarge	32800	45200	+37.8%	0.73x (?)

Performance: -Osize

TEST	OLD	NEW	DELTA	RATIO
Regression
DataAppendDataLargeToLarge	33000	45400	+37.6%	0.73x (?)
ObjectiveCBridgeStubFromNSDateRef	3640	3940	+8.2%	0.92x (?)

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

[Valpertui]

Maybe we should add some tests to ensure those kind of compile time regression doesn't get released unnoticed anymore ?