allow showProgress=INTEGER to set progress bar update time

[ben-schwen]

Closes #6514

# simple example to trigger progress bar
f = function(x) {Sys.sleep(1); x}
dt = data.table(x = 1:10, y = 1:10)
dt[, f(x), y, showProgress=3L]

[codecov]

Codecov Report

❌ Patch coverage is 50.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 98.79%. Comparing base (e2e0173) to head (3990670).
⚠️ Report is 6 commits behind head on master.

Files with missing lines	Patch %	Lines
R/data.table.R	50.00%	1 Missing ⚠️
src/dogroups.c	50.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #7259   +/-   ##
=======================================
  Coverage   98.79%   98.79%           
=======================================
  Files          81       81           
  Lines       15255    15255           
=======================================
  Hits        15071    15071           
  Misses        184      184           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

No obvious timing issues in HEAD=showProgress_update

Generated via commit 3990670

Download link for the artifact containing the test results: ↓ atime-results.zip

Task	Duration
R setup and installing dependencies	4 minutes and 53 seconds
Installing different package versions	10 minutes and 25 seconds
Running and plotting the test cases	2 minutes and 48 seconds

[tdhock]

Thanks for the PR Ben. Is there a reason why the delay is integer rather than double/numeric? (which seems more natural for a time)

[ben-schwen]

Thanks for the PR Ben. Is there a reason why the delay is integer rather than double/numeric? (which seems more natural for a time)

I also thought about using doubles first, but then integer seemed like the right choice since everything below 1/4 second is hard to grasp so the granularity seems to be coarse enough and I also didn't want to cater for strange supplied values like NA (behaves like -INT_MAX), Inf, etc.