Address review comments

Merge from Widevine repo of http://go/wvgerrit/122383

Fix a few grammar errors found in http://go/ag/14154405 and
http://go/ag/14150103.

Bug: 185536454
Test: comments changed only
Change-Id: Ied2af940d705fbfc4af60c1d76a051c8c163097f

libwvdrmengine/cdm/core/test/duration_use_case_test.cpp

@@ -1055,7 +1055,7 @@ class RenewalTest : public CdmDurationTest {
     std::string http_response;
     // Most of the network latency will probably show up in the next few
     // commands. I think the tests have enough slop to account for reasonable
-    // latency with the current value of kRoundTripTime. But We'll know I made a
+    // latency with the current value of kRoundTripTime. But we'll know I made a
     // mistake if we see errors about "Test Clock skew..." in the SleepUntil
     // call below.
     ASSERT_TRUE(renewal_in_flight_->GetResponse(&http_response));

libwvdrmengine/oemcrypto/test/README.md

@@ -81,11 +81,11 @@ the ODK functions as described in the document "License Duration and Renewal".
 
 ### What to expect from these tests
 
-*   `OEMCryptoMemory*` tests are designed to fail if API doesn't have enough
-    validations around input buffer lengths, parameters or OEMCryptoSubstring
-    struct. If the API doesn't have validations which might lead to a crash, the
-    test fails with a segfault or an appropriate crash message based on the API
-    implementation.
+*   `OEMCryptoMemory*` tests are designed to fail if the API doesn't have
+    enough validations around input buffer lengths, parameters or the
+    OEMCryptoSubstring struct. If the API doesn't have validations which might
+    lead to a crash, the test fails with a segfault or an appropriate crash
+    message based on the API implementation.
 
 *   Find out for what buffer length, the API is crashing and then debugging the
     test against the OEMCrypto implementation should be able to provide