Rephrase instructions for local usage

[timholy]

The rewords some of the instructions in the README for local coverage measurement. The major changes & their motivations are:

• indicate the purpose of each step before describing the mechanics of the step. This also makes it easier to indicate that one does not necessarily need to measure summary statistics if one's only goal is to find lines lacking coverage.
• prioritize Pkg.test(; coverage=true) above manual measurement. Most users will want the simpler option.
• clarify the role of the .info tracefile. This allows users to determine when they need to supply the more complex set of options. By presenting it as a choice it also makes it clearer that you don't need to execute both lines.
• acknowledge limitations in the accuracy of Julia's coverage analysis

[coveralls]

Pull Request Test Coverage Report for Build 1345843085

• 0 of 0 changed or added relevant lines in 0 files are covered.
• 4 unchanged lines in 2 files lost coverage.
• Overall coverage increased (+0.5%) to 90.865%

---

Files with Coverage Reduction	New Missed Lines	%
src/codecovio.jl	2	87.38%
src/coveralls.jl	2	94.29%
<!--	Total:	4		-->

Totals
Change from base Build 1308404670:	0.5%
Covered Lines:	189
Relevant Lines:	208

---

💛 - Coveralls

[codecov[bot]]

Codecov Report

Merging #327 (a7a3088) into master (c3fe9bc) will increase coverage by 0.47%. The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #327      +/-   ##
==========================================
+ Coverage   90.09%   90.56%   +0.47%     
==========================================
  Files           2        2              
  Lines         212      212              
==========================================
+ Hits          191      192       +1     
+ Misses         21       20       -1     

Impacted Files	Coverage Δ
src/coveralls.jl	94.28% <0.00%> (-1.91%)	:arrow_down:
src/codecovio.jl	86.91% <0.00%> (+2.80%)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update c3fe9bc...a7a3088. Read the comment docs.

[timholy]

This is for a class I'm teaching, and the homework gets assigned today, so I'll merge.