[SIL.WIndows.Forms] In CustomDropDown, fixed timer check

[imnasnainaec]

If you start with double 0 and add .1 ten times, you can end up with (e.g.) 0.9999999999999999 instead of 1f.

This fixes https://github.com/sillsdev/libpalaso/security/code-scanning/2627

Of the other 14 warnings for "Equality check on floating point values" (https://github.com/sillsdev/libpalaso/security/code-scanning?query=is%3Aopen+branch%3Amaster+rule%3Acs%2Fequality-on-floats), three involve screen DPI (2624, 2625, 2626), and the rest involve font size.

---

This change is 

[github-actions]

Palaso Tests

     4 files  ±0       4 suites  ±0   11m 10s ⏱️ +55s
 4 978 tests ±0   4 743 ✅  - 1  235 💤 +1  0 ❌ ±0 
16 186 runs  ±0  15 481 ✅  - 2  705 💤 +2  0 ❌ ±0 

Results for commit f957a4b. ± Comparison against base commit 23dc405.

This pull request skips 1 test.

SIL.Tests.IO.FileLocationUtilitiesTests ‑ LocateInProgramFiles_SendValidProgramDeepSearch_SubFolderSpecified_ReturnsProgramPath

♻️ This comment has been updated with latest results.

[tombogle]

Reviewed 1 of 2 files at r2, all commit messages.
Reviewable status: 1 of 2 files reviewed, 1 unresolved discussion (waiting on @imnasnainaec)

---

SIL.Windows.Forms/Widgets/CustomDropDown.cs line 187 at r2 (raw file):

			{
				Opacity += 0.1;
				if (Opacity > .99)

Should we also actually force opacity to be 1f? In practice, I imagine it won't make a noticeable difference (maybe not any difference), so maybe this is just me feeling angst over the imprecision.

[tombogle]

Reviewed 1 of 2 files at r2, all commit messages.
Reviewable status: 1 of 2 files reviewed, 1 unresolved discussion (waiting on @imnasnainaec)

[imnasnainaec]

Reviewable status: 1 of 2 files reviewed, 1 unresolved discussion (waiting on @tombogle)

[tombogle]

Reviewed 1 of 1 files at r4, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @imnasnainaec)