Button: Perf improvements

[khmakoto]

…le, small perf improvements and perf scenario updates.

Pull request checklist

• Addresses an existing issue: Fixes #0000
• Include a change request file using $ yarn change

Description of changes

This PR creates a separate Actionable view that speeds up the Actionable component because it gets rid of unused slots and spreads some small perf fixes that check for the existence of icon and content before rendering them (these are not reflected in the perf-test scenarios).

Finally, this PR updates perf-test scenarios for Actionable and PrimaryButton so that they are more consistent across the board.

Focus areas to test

(optional)

Microsoft Reviewers: Open in CodeFlow

[size-auditor]

Bundle test	Size (minified)	Diff from master
experiments-Button	94.789 kB	723 bytes

  Exceeds Tolerance       Exceeds Baseline       Below Baseline     1 kB = 1000 bytes

[msft-github-bot]

Component Perf Analysis:

Scenario	Master Samples *	PR Samples *
BaseButton	732	765
BaseButton (experiments)	1690	1074
DefaultButton	1020	1068
DefaultButton (experiments)	1919	1977
DetailsRow	4096	4205
DetailsRow without styles	3954	3940
DocumentCardTitle with truncation	30919	30343
MenuButton	1806	1829
MenuButton (experiments)	4237	4046
PrimaryButton	1152	1192
PrimaryButton (experiments)	2201	1937
SplitButton	3297	3284
SplitButton (experiments)	7783	7404
Stack	447	474
Stack with Intrinsic children	1046	1098
Stack with Text children	4183	4160
Text	358	342
Toggle	829	841
Toggle (experiments)	2149	2379
button	57	56

* Sample counts can vary by up to 30% and shouldn't be used solely for determining regression. For more information please see the Perf Testing wiki.

[msft-github-bot]

🎉@uifabric/[email protected] has been released which incorporates this pull request.:tada:

Handy links:

• Release Notes
• Official NPM Release